本篇博文主要内容为 2026-09-23 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 (2026-09-23)
今日共更新918篇论文,其中:
- 自然语言处理共110篇(Computation and Language (cs.CL))
- 人工智能共243篇(Artificial Intelligence (cs.AI))
- 计算机视觉共143篇(Computer Vision and Pattern Recognition (cs.CV))
- 机器学习共218篇(Machine Learning (cs.LG))
- 多智能体系统共15篇(Multiagent Systems (cs.MA))
- 信息检索共16篇(Information Retrieval (cs.IR))
- 人机交互共31篇(Human-Computer Interaction (cs.HC))
多智能体系统
[MA-0] Agensh: Scaling Organizational Intelligence to 1024 Agents
【速读】：该论文旨在解决现有多智能体（multi-agent）调度框架在扩展性方面受限于中心协调器（central orchestrator）任务分配与工作流协调能力的问题。传统架构中，中心化设计成为系统吞吐量和并行效率的瓶颈，尤其在处理复杂、高延迟敏感的任务时表现尤为明显。为突破这一限制，本文提出Agensh——一种无需中心协调器的可扩展自组织多智能体调度框架。其核心解决方案在于构建一个去中心化的智能体协作机制：多个并发执行的智能体通过异步循环持续完成“获取上下文—主动认领并自主分配子任务—采取行动—共享成果—验证结果—合并进展”的全流程协作。该协作循环依托三大基础设施支撑：共享工作空间（用于存放待办、进行中及已完成的工作）、消息接口（实现智能体间通信）以及共享上下文（存储可复用的发现与工作意图）。实验表明，在ProgramBench五个最困难的任务上，当代理数量从1增加至128时，平均测试通过率由19.31%提升至28.78%，相对提升约49%；在pandoc任务中，从1到1024个代理的扩展使通过率从33.89%增至55.06%。此外，工作轨迹分析显示，随着组织规模扩大，不同形式的自组织协作模式逐渐涌现并趋于标准化。这些结果揭示了智能体数量本身可作为拓展通用智能边界的新型可扩展维度，为在严苛延迟约束或时间预算下执行复杂任务提供了高效可行的解决方案。
链接: https://arxiv.org/abs/2609.26781
作者: Zhihao Zhan,Ting Song,Li Dong,Shaohan Huang,Jianxun Lian,Yan Xia,Furu Wei
机构: Microsoft Research(微软研究院)
类目: Computation and Language (cs.CL); Multiagent Systems (cs.MA)
备注: 13 pages, 6 figures
Abstract:A multi-agent system can reduce latency on complex tasks by executing work concurrently. Several pioneering harness frameworks support multi-agent systems. However, the scalability of current multi-agent harnesses is often constrained by a central orchestrator’s capacity to allocate tasks and coordinate workers. To address this limitation, we introduce Agensh, a scalable self-organized multi-agent harness without a central orchestrator: concurrent workers execute a multi-agent cooperation loop, continuously gathering context, claiming and self-assigning sub-tasks, taking action and sharing findings, verifying results, and merging progress in an asynchronous manner. The loop is supported by the agentic organization infrastructure comprising three components: a shared workspace holds proposed, ongoing, and completed work; a message interface lets workers communicate; and shared context retains reusable findings and work intentions. To test the scalability of Agensh, we evaluate it on the five hardest ProgramBench tasks with GPT-5.6-sol (high). Scaling from 1 to 128 agents raises the mean final test-pass rate from 19.31% to 28.78%, an approximately 49% relative improvement. Larger organizations reach comparable test-pass rates earlier. On pandoc, scaling from 1 to 1,024 agents raises the final test-pass rate from 33.89% to 55.06%. Worker trajectories further show that different forms of self-organized cooperation gradually emerges and standardizes as the organization grows. These results reveal the number of agents as a new scaling dimension for multi-agent organizations to expand the frontier of general intelligence, offering a practical solution for complex tasks under hard latency constraints or time budgets.
[MA-1] Behavior is Not Enough: A Mechanism-Based Evaluation of Social Norm Emergence in LLM Societies AAAI2027
【速读】：该论文旨在解决多智能体大语言模型（multi-agent large language model）系统中，仅依据行为趋同便推断社会规范（social norm）形成的局限性问题。传统方法将行为收敛视为规范涌现的证据，但相同的合作均衡可能源于共享预期、策略激励或简单模仿等不同机制，难以区分。为此，论文提出一种评估框架，不仅测量行为收敛，还通过显式采集智能体的实证预期（empirical expectations）与规范性预期（normative expectations），以识别规范形成背后的深层社会过程。其解决方案的关键在于：通过可控消融实验，分离出两种核心集体机制——基于交互的社会学习（social learning through interaction）与基于网络的群体形成带来的社会选择（social selection through network-based group formation）。研究发现，显式诱导预期可提升合作贡献，社会学习有助于稳定行为模式，而社会选择能有效识别合作者但对行为强化作用有限；在对抗性干扰下，规范性预期与行为协调的恢复路径不同。结果表明，相似的合作成果可能由不同社会机制驱动。该框架通过使预期可观测，实现了对各机制贡献的独立归因，为多智能体系统设计者提供了基于原则的选择依据，以构建可持续合作的社会过程。
链接: https://arxiv.org/abs/2609.26481
作者: Rasika Muralidharan,Haewoon Kwak,Jisun An
机构: 未知
类目: Multiagent Systems (cs.MA); Computation and Language (cs.CL); Computers and Society (cs.CY); Computer Science and Game Theory (cs.GT); Social and Information Networks (cs.SI)
备注: Under review at AAAI 2027 Special Track: AI Alignment
Abstract:Social norms cannot be identified from behavior alone: the same cooperative equilibrium may reflect shared expectations, strategic incentives, or simple imitation. Yet in multi-agent large language model systems, prior work largely treats behavioral convergence as evidence of norm emergence. In this work, we introduce an evaluation framework that measures agents’ reported empirical and normative expectations in addition to behavioral convergence. Through controlled ablations, we test the effect of expectation elicitation and isolate two collective mechanisms central to theories of norm formation—social learning through interaction and social selection through network-based group formation. We further test the stability of these resulting dynamics under adversarial disruption across four LLM families. We find that eliciting expectations increases cooperative contributions, while social learning stabilizes behavior, and social selection reliably identifies cooperators but provides limited behavioral reinforcement. Following disruption, normative expectations and behavioral coordination recover differently. Together, these results show that similar cooperative outcomes can arise from different underlying social processes. By making expectations observable, our framework allows us to attribute each mechanism’s contribution separately, offering designers of multi-agent systems a principled basis for selecting the social processes that sustain cooperation.
[MA-2] From Risk Scoring to Risk Allocation: A Density-Driven Framework for Diverse Monitoring in Multi-Agent Systems ICML2026
【速读】：该论文旨在解决多智能体系统中风险监控因“聚集效应”（crowding）导致的冗余告警问题，即传统基于单状态独立评分的方法在多个智能体具有相似脆弱性时，会频繁选择风险高度相关的状态，形成“监控中的羊群效应”（herding in monitoring）。其核心解决方案是实现从风险评分到风险分配的范式转变，关键在于提出并验证了“聚集悖论”（Crowding Paradox），揭示出条件风险概率 P(risk∣x) 与状态密度 p(x) 成正比，而非与异常分数成反比，表明密度才是真正的风险信号。基于此，研究将风险监控重构为具有相互依赖性的状态组合选择问题，并将其建模为带 λ 控制的风险-多样性权衡的二次无约束二值优化（QUBO）问题。该框架的帕累托前沿包含标准多样化子集方法（如MMR、k-DPP）作为固定操作点，且相较于贪婪算法的性能增益随规模增加而单调提升（从 n=15 时的 +24% 到 n=200 时的 +66%），通过学习 λ 策略可达到接近全网格搜索最优解的 99.5%，且该模型在交通系统和多智能体强化学习中具备良好可迁移性。此外，相同 QUBO 实例可在 Rigetti 超导量子处理器（Ankaa-3 与 Cepheus-1-108Q）上直接执行，验证了该公式的兼容性，但不构成在此规模下的量子优势主张。
链接: https://arxiv.org/abs/2609.26146
作者: Zhaohui Wang
机构: 未知
类目: Machine Learning (cs.LG); Multiagent Systems (cs.MA)
备注: 13 pages, 7 figures. Accepted at the ICML 2026 Workshop on New Frontiers in Game-Theoretic Learning (NExT-Game)
Abstract:Risk monitoring in multi-agent systems is commonly built on a per-state primitive that scores each state independently and selects the top K. Under crowding, where many agents share the same fragility, this approach picks redundant alerts whose risks are jointly correlated, a pattern we describe as ``herding in monitoring.‘’ We propose a paradigm shift from risk scoring to risk allocation, supported by two contributions. First, we identify the Crowding Paradox, namely that P(risk | x) \propto p(x) rather than 1/p(x), so density rather than anomaly score is the operative risk signal; on financial data, density-based scoring reaches AUROC \geq 0.94 at 5d/10d/20d crash horizons, while five anomaly baselines all fall below 0.80. Second, given a density-derived fragility score, we recast monitoring as combinatorial subset selection over interdependent states and map it to a QUBO objective with a \lambda -controlled risk–diversity tradeoff. The resulting Pareto frontier contains standard diverse-subset methods (MMR, k-DPP) as fixed operating points; the gain over greedy grows monotonically with scale, from +24% at n=15 to +66% at n=200; a learned \lambda policy reaches 99.5% of an oracle grid-search objective; and the formulation transfers to traffic and multi-agent reinforcement learning. The same QUBO instances execute without modification on Rigetti superconducting QPUs (Ankaa-3 and Cepheus-1-108Q via Amazon Braket), which we report as a compatibility property of the formulation rather than a claim of quantum advantage at this scale.
[MA-3] VACS: Value-Aligned Compositional Shielding for Multi-Agent Reasoning
【速读】：该论文旨在解决高风险领域中多智能体推理系统在准确性与安全性之间难以兼顾的核心问题，尤其针对智能体间因异质价值优先级（如严谨性、简洁性、安全性）差异导致的推荐冲突。现有方法无法同时实现：(i) 从行为中可解释地推断各智能体的隐含价值；(ii) 在不依赖全程在线通信的前提下提供组合式的形式化安全保证；(iii) 基于价值感知的冲突消解并生成可信的解释。其解决方案的关键在于提出VACS（Value-Aligned Compositional Shielding）四层框架：第一层通过Bradley-Terry建模与深度最大熵逆强化学习（MaxEnt IRL），从成对偏好数据中学习价值维度奖励并推断各智能体的价值权重；第二层采用受Lean启发的领域特定语言（DSL）编码价值约束，并合成组合式假设-保证防护机制以保障运行时安全；第三层基于核心值分配（nucleolus-based credit allocation）与哈密顿共识优化，在长期价值约束下实现冲突消解；第四层利用共态敏感性提取关键推理路径，并生成形式化支撑的自然语言解释。整体贡献在于构建了一个统一的系统架构，具备形式化接口与验证器约束下的决策级操作保障，而非对语言模型内部进行完整端到端的形式化证明。在NEJM-AI问答、MathInstruct-Subset及网络安全事件响应基准（CyberSec-Eval）上的对照实验表明，VACS在准确率（85.4%、95.0%、90.0%）上超越强基线，同时将逻辑不一致率降至接近零。
链接: https://arxiv.org/abs/2609.26135
作者: Yiyao Zhang,Diksha Goel,Hussain Ahmad,Shixun Huang,Jun Shen
机构: University of Wollongong (伍伦贡大学); CSIRO’s Data61 (澳大利亚联邦科学与工业研究组织数据61实验室); Adelaide University (阿德莱德大学)
类目: Artificial Intelligence (cs.AI); Multiagent Systems (cs.MA)
备注:
Abstract:Multi-agent reasoning systems in high-stakes domains must be both accurate and safe, yet agents often follow heterogeneous value priorities (e.g., rigor, conciseness, safety), causing conflicting recommendations. Existing methods do not jointly provide: (i) principled inference of each agent’s implicit values from behavior, (ii) compositional formal safety guarantees without full online communication, and (iii) value-aware conflict resolution with faithful explanations. We present VACS (Value-Aligned Compositional Shielding), a four-layer framework addressing all three. Layer 1 learns value-dimension rewards from pairwise preferences using Bradley-Terry modeling and infers per-agent value weights via deep MaxEnt IRL. Layer 2 encodes value constraints in a Lean-inspired DSL and synthesizes compositional assume-guarantee shields for runtime safety. Layer 3 resolves disagreement through nucleolus-based credit allocation and Hamiltonian consensus optimization under long-term value constraints. Layer 4 extracts a critical reasoning path from co-state sensitivities and generates formally grounded natural-language explanations. Our contribution is primarily a unified systems design with formalized interfaces and operational guarantees at the verifier-constrained decision level, rather than a complete end-to-end formal proof of all language-model internals. In controlled proof-of-concept evaluations with role-conditioned agent panels on NEJM-AI QA, MathInstruct-Subset, and a cybersecurity incident-response benchmark (CyberSec-Eval), VACS outperforms strong baselines in accuracy (85.4%, 95.0%, and 90.0%) while reducing logical inconsistency rates to near zero.
[MA-4] DTOC: Dynamic Tool Output Compression for Adaptive Context Management in AI Agents
【速读】：该论文旨在解决大语言模型（Large Language Model, LLM）代理在复杂任务推理过程中因上下文窗口（context window）受限而导致的信息丢失与效率下降问题。传统方法如截断、启发式老化和有损摘要，虽能缓解上下文压力，但易造成有用信息丢失或引入幻觉风险。为此，论文提出动态工具输出压缩（Dynamic Tool Output Compression, DTOC）框架，其核心在于将上下文更新建模为代理推理循环中显式且可逆的操作：将完整的工具输出存储于外部记忆中，仅在活跃上下文中插入紧凑的占位符，并支持按需重构。该机制通过形式化定义与基于ReAct架构的集成，实现了生产级的按需恢复能力。实验结果表明，在DeepSWE基准测试中，对于响应型模型（Sonnet 4.6、GPT-5.4），DTOC显著降低输入令牌数（分别减少10.3%和12.7%）、代理步骤（分别减少2.4%和32.3%），提升求解率（分别提高2.5倍和1.5倍），并使每成功任务成本降低3倍和3.5倍；而对于其他模型则表现各异，部分模型虽提升求解率并降低成本，但亦存在性能退化情况。消融实验进一步验证了可逆性的重要性：仅压缩而不可逆的变体导致性能下降，而完整DTOC在大幅降低上下文开销的同时恢复了基线准确率。研究结论表明，显式且可逆的上下文管理机制可在不损害任务性能的前提下，有效提升长时程代理推理的效率。
链接: https://arxiv.org/abs/2609.26121
作者: Abhay Chaturvedi,Shreya Bhattacharya,Rashmika Gopalkrishnan,Peter van der Putten
机构: 未知
类目: Artificial Intelligence (cs.AI); Computation and Language (cs.CL); Multiagent Systems (cs.MA)
备注: Accepted for 29th International Conference on Discovery Science, October 5-9, 2026, Mainz, Germany
Abstract:As agent capabilities have grown, practical limitations increasingly stem from constrained context windows rather than model capacity. Common strategies, such as truncation, heuristic aging, and lossy summarization, may discard useful information or introduce hallucination risk. To address these challenges, we propose Dynamic Tool Output Compression (DTOC), a framework for scalable context management in LLM-based agents that models context updates as explicit and reversible operations within the agent reasoning loop. DTOC retains full tool outputs in external memory while inserting compact placeholders into the active context, enabling selective reconstruction when needed. We formalize the DTOC mechanism, integrate it into a ReAct-style agent architecture, and provide a production-oriented implementation supporting on-demand restoration of compressed outputs. Experiments on DeepSWE reveal model-dependent effects: for responsive models (Sonnet 4.6, GPT-5.4), DTOC reduces input tokens (10.3 and 12.7%) and agent steps (2.4 and 32.3%), while increasing solve rates (2.5 and 1.5 times higher) and lowering cost per solved task (3 and 3.5 times lower cost per solved task). For the other models results are more mixed, with GPT-5.5 doubling solve rate and halving cost, but no impact on solve rate and negative impact on cost for the other models. Ablation results show reversibility is critical: disable-only compression variants degraded performance, while full DTOC recovered baseline accuracy at substantially lower context cost. These findings indicate that explicit, reversible context management can improve the efficiency of long-horizon agent reasoning without degrading task performance.
[MA-5] he Fleet Is the Model: Engineering Collective Intelligence with Fusion-MoA Pioneer R1
【速读】：该论文旨在解决多模型协同服务中缺乏统一身份标识、权限管理与成员演化能力的问题，即现有推理系统虽能管理模型检查点与副本，但多智能体框架在组合模型调用时无法定义稳定的集体身份、权限边界或个体演化机制。其核心解决方案是提出Fusion-MoA运行时系统，将独立部署的异构模型单元（Cells）整合为一个兼容OpenAI接口的统一服务。该系统通过版本化配置文件（Profile）控制成员资格与证据接纳规则，引入只读分析者（Analysts）提供有限证据输入，而单一执行者（Executor）保留最终答案生成与工具调用的唯一权威。此设计实现了模型单元的动态升降级与回滚而不影响对外API，且在多个评估场景中验证了其有效性：在固定测试集上，集体表现优于最强单个单元，知识可跨单元迁移；在终端任务测试中，所有工具操作可追溯至单一执行者，无越权行为或绕过效应；同时支持六次单元晋升与一次不兼容单元本地回滚，服务持续可用。结果表明，异构模型能力可被作为可观测、权限受限且可独立演进的服务进行统一调度与管理。
链接: https://arxiv.org/abs/2609.26080
作者: Zongyou Yang,Yinghan Hou
机构: 未知
类目: Multiagent Systems (cs.MA); Distributed, Parallel, and Cluster Computing (cs.DC)
备注: 9 pages, 3 figures, and 5 tables
Abstract:The model exposed to an application need not be a single checkpoint; it can be a governed fleet. Existing serving systems manage checkpoints and replicas, while multi-agent frameworks compose model calls without defining a stable collective identity, effect authority, or member-level evolution. We present Fusion-MoA, a runtime that exposes independently served heterogeneous Cells as one OpenAI-compatible model. A versioned Profile controls membership and evidence admission; read-only Analysts contribute bounded evidence, while a sole Executor retains all final-answer and tool authority. Cells can be qualified, promoted, or rolled back without changing the public API. We evaluate an eight-Cell, three-lineage deployment through three operational witnesses. On a fixed HMMT P1-P10 slice, the collective solves 8/10 problems versus 6/10 for the strongest individual Cell, and a preserved trace shows minority knowledge transferred to three initially incorrect or empty Cells. On 20 Terminal-Bench 2.1 tasks, all tool actions remain attributable to one Executor, with zero Analyst actions and zero bypass effects. Six Cells are promoted and one incompatible candidate is locally rolled back while the service remains available. Fusion-MoA demonstrates that heterogeneous model capability can be operated as one observable, authority-bounded, and independently evolvable service.
[MA-6] Adversarial Course-of-Action Generation: Game-Theoretic Multi-Agent Algorithms for COA matching COA generation
【速读】：该论文旨在解决生成式战术规划中行动方案（Course-of-Action, COA）生成的分布式协同规划问题，即如何在对抗性环境下游走于动态变化条件下，自动生成结构化候选行动方案，并评估其在面对敌方反制时的战术一致性与有效性。其核心挑战在于确保生成的COA不仅具备战术合理性，还能在多智能体对抗博弈中保持稳健性与可解释性。解决方案的关键在于提出一个名为COA-Bench的小规模离线基准测试框架，作为可复现的评估工具，用于比较不同COA生成策略在自我对弈（self-play）场景下的表现。该框架将COA表示为带有条件分支的类型化动作链，引入合成的COA质量评分、基于蓝红双方优势差（BLUE-vs-RED advantage score）与纳什差距（Nash-gap distance）的对抗评估指标，以及受FM 3-0作战条令启发的教义一致性评分规则。通过50个覆盖五种作战模板的合成场景验证，结果表明：采用采样最优响应策略可显著降低蓝方优势（从0.516降至0.485）并削弱胜率（从0.920降至0.820），而采用双阶段多智能体协商机制（含五名蓝方提案代理、红方评估与批判性修订）则实现更优平衡（蓝方优势0.509，胜率0.820）。此外，研究还识别并修正了基准设计缺陷——场景框架信息虽被存储为元数据但未影响生成内容，提升了评估可信度。整体贡献在于提供了一个可审查、可复现的评估架构，初步验证了生成式规划系统的可行性，并为构建可审计的智能体规划系统提供了关键实践启示。
链接: https://arxiv.org/abs/2609.26059
作者: Natan Vidra,Alina Kapanova,Arun Kanhai,Spurthi Setty
机构: Anote AI(安特奥人工智能); Cornell University (康奈尔大学); CUNY (纽约市立大学); Stevens Institute of Technology (史蒂文斯理工学院)
类目: Artificial Intelligence (cs.AI); Multiagent Systems (cs.MA)
备注: 5 pages. DAI 2026 Industry Track submission. Includes public reproducibility artifact link. Uses synthetic offline scenarios only; no classified data, proprietary data, human-subject data, or operational planning systems
Abstract:Course-of-action (COA) generation is a distributed planning problem: a system must propose structured candidate actions, evaluate them against an adversarial response, and surface options that remain tactically coherent under changing conditions. We present COA-Bench, a small offline benchmark and reproducibility artifact for comparing COA generation policies through self-play. Following the BattleCOA terminology, we reserve COA matching for asset-effect matching and COA generation for course-of-action generation; the present artifact does not implement either DecisionFunction directly. Instead, it represents COAs as typed action chains with conditional branches, assigns a synthetic COA quality score, compares opposing COAs with a BLUE-vs-RED advantage score and Nash-gap distance, and scores doctrinal coherence with an FM 3-0-inspired heuristic rubric. Across 50 synthetic scenarios spanning five operational templates, a sampled best-response policy that draws eight RED candidates reduces BLUE advantage from .516 to .485 and BLUE wargame win rate from .920 to .820; a two-stage multi-agent council with five BLUE proposer agents, RED-team adjudication, and critique-driven revision obtains .509 BLUE advantage and .820 BLUE win rate. We also identify and fix a benchmark-design issue in which scenario framing was stored as metadata but had no effect on generated COA content. COA-Bench is not an operational battle-management system and uses no real, classified, proprietary, or human-subject data. The contribution is an inspectable evaluation harness, preliminary benchmark evidence, and lessons for building auditable agentic planning artifacts.
[MA-7] MATES: Learning Multi-Agent Interactions by Transforming Observations for Frozen Single-Agent Policies
【速读】：该论文旨在解决多智能体强化学习（MARL）中分布式策略从零开始训练时，智能体需同时习得个体任务能力与协作机制所带来的高训练成本与复杂性问题。其核心挑战在于如何在不修改预训练单智能体策略的前提下，高效地将其适配至多智能体协作场景。解决方案的关键是提出一种基于输入端适配的框架——多智能体观测变换（MATES），该框架利用多智能体观测中保留的独立任务信息与可区分的邻近智能体信息结构，通过学习一个轻量级适配器（adapter），将多智能体观测映射为预训练单智能体策略所期望的输入格式。该适配器仅优化少量参数（仅为全策略训练的3.5%-7.3%），且不更新预训练策略的内部架构或底层MARL算法的优化目标，从而实现对个体任务能力的保留与协作行为的有效诱导。实验表明，MATES在长期路径规划、导航及协同发现等任务中显著优于从零开始的MARL训练，性能接近全参数微调，并在未见团队规模下仍保持强泛化能力，验证了在特定观测结构下，无需修改策略主体即可实现高效多智能体行为学习的可行性。
链接: https://arxiv.org/abs/2609.26010
作者: Elie Abboud,Oren Gal
机构: University of Haifa(海法大学); Department of Marine Technologies(海洋技术系)
类目: Multiagent Systems (cs.MA); Robotics (cs.RO); Systems and Control (eess.SY)
备注:
Abstract:Multi-agent reinforcement learning (MARL) commonly trains decentralized policies from scratch, requiring agents to acquire individual task competence and coordination simultaneously. Yet many multi-agent problems admit a compatible single-agent counterpart in which the underlying task can be learned in isolation. We introduce Multi-Agent Observation Transformation for Existing Single-Agent Policies (MATES), an input-side adaptation framework for tasks whose multi-agent observations preserve the solo-task information while exposing separately identifiable neighbor information. From multi-agent experience, MATES learns a small adapter that maps this observation into the format expected by a frozen single-agent policy, inducing actions suited to the shared environment without updating the single-agent policy itself. MATES leaves the pretrained policy’s internal architecture unchanged and retains the objectives and update procedures of the underlying MARL algorithm. We evaluate MATES using both on- and off-policy algorithms on lifelong pathfinding, navigation, and cooperative discovery, spanning discrete and continuous observation and action spaces. Across all evaluated settings, MATES optimizes only 3.5-7.3% as many parameters as full-policy training while consistently outperforming MARL training from scratch. It approaches the performance of full fine-tuning, remains competitive overall with demonstration-based baselines, and retains strong task performance at team sizes not encountered during training. These results provide evidence that, under this observation structure, effective multi-agent behavior can be learned without modifying the policy that encodes individual competence.
[MA-8] Calibration Is Not Verification: Falsifiability-Aware Conformal Routing for Mixture-of-Agents
【速读】：该论文旨在解决多智能体语言系统（Multi-agent language systems）在生成长文本时因依赖共识而产生的事实性不可靠问题，即异质智能体可能共同重复一个无支持的错误主张或遗漏正确专家知识。其核心解决方案是提出C-MoA（Consensus-based Mixture-of-Agents），一种基于协议的置信度校准过滤器，将智能体间的语义支持转化为声明级别的非符合度（nonconformity）分数，并在示例层面校准保留阈值，从而实现无需分布假设的域内事实性控制。关键创新在于通过置信度校准实现可迁移、可靠的精度提升：在长文本生成中将保留声明的精确率从0.41显著提升至0.75，且在无需重新校准的情况下跨领域有效，但在短文本回答场景下因共识成本过低而表现受限。为进一步突破共识局限，论文进一步提出CONTRA-MoA，引入盲化近似失败竞赛（blinded near-miss tournament）、逐智体剔除稳定性检验（leave-one-agent-out stability）以及可用性感知融合机制，实验证明该方法仅在验证者具备领域知识时有效，可将假医疗声明减少一半并达到0.940的高精确率；而在仅依赖记忆的判断者情况下，新增信号接近随机水平（AUC分别为0.531和0.511），且朴素的最大融合策略反而削弱了原有共识信号（从0.687降至0.652）。研究结论表明：基于共识的置信度校准可提供可靠且可迁移的事实性保障，但超越共识必须依赖具备领域知识的验证者、考虑可用性的信号设计及鲁棒融合机制。
链接: https://arxiv.org/abs/2609.25959
作者: Nada Rahali,Zijia Wang,Zhisong Liu
机构: 未知
类目: Multiagent Systems (cs.MA)
备注: 14 pages, 5 figures
Abstract:Multi-agent language systems often treat agreement as evidence, yet heterogeneous agents can jointly repeat an unsupported claim or omit a correct specialist fact. We introduce C-MoA, an agreement-based conformal filter that turns inter-agent semantic support into a claim-level nonconformity score and calibrates a retention threshold at the example level, giving distribution-free within-domain factuality control for heterogeneous Mixture-of-Agents. C-MoA is effective: it nearly doubles retained-claim precision on long-form generation (from 0.41 to 0.75), certifies a human-labelled medical set, and transfers across domains without recalibration; its one failure mode is short-form answering, where consensus is cheap and the score is left near chance. We then ask whether counterfactual falsifiability can push past consensus, and introduce CONTRA-MoA, which adds a blinded near-miss tournament, leave-one-agent-out stability, and availability-aware fusion. This extension helps only where the verifier holds domain knowledge, dropping half of the false medical claims at 0.940 precision, whereas with a memory-only judge the added signals are near chance (AUC 0.531 and 0.511) and naive max fusion degrades the working agreement signal from 0.687 to 0.652. The message is twofold: agreement-based conformal calibration delivers reliable, transferable factuality control, while moving beyond consensus requires a knowledgeable verifier, availability-aware signals, and robust fusion.
[MA-9] Governed AI-Agent Coordination for Dementia Care: Architecture Safety Contracts and Evidence-Derived Workflow Verification
【速读】：该论文旨在解决社区痴呆症照护中多系统异构集成所导致的协作失效问题，核心挑战在于现有互操作性机制无法维持可问责的照护状态、协调证据、明确行动权限或验证处置结果。其解决方案的关键是提出一种受控闭环代理协同架构（Governed Closed-loop Agent Coordination, GCAC），通过引入具有外部运行时能力的代理工程范式，实现跨会话的记忆保持、基于目标与约束的规划、工具调用、结果观测及治理规则强制执行。GCAC采用类型化事件-记忆-决策-动作-结果契约，将观察、受控记忆、规划、确定性策略执行、执行与结果监控等环节解耦，并以18个基于临床证据的测试轨迹验证其有效性，涵盖缺失记录、用药冲突、照护者报告异常、服务失败、授权变更、过期状态、重复事件、不可信文本及疑似急性神经功能变化等典型场景。实验表明，GCAC在所有18个契约断言上均满足要求，零次违反策略的工具调用，准确保留照护义务、拒绝过期状态、触发人工交接并记录工作流关闭；而事件阈值与无状态规划器控制仅分别满足2/18和1/18断言，组件消融分析进一步定位了故障根源至记忆、策略或版本管理模块。研究结果证明了该架构的系统符合性，而非临床疗效，展示了代理系统在自动化证据重整、任务路由、文档生成与随访方面的能力，同时确保关键照护决策权始终保留在人类主体手中。
链接: https://arxiv.org/abs/2609.25956
作者: Francesca Medda,Hui Gong
机构: UCL Institute of Finance and Technology (伦敦大学学院金融科技研究所); University College London (伦敦大学学院); London, United Kingdom
类目: Multiagent Systems (cs.MA)
备注: 13 pages, 2 figures, 5 tables. Includes an executable architecture-verification harness with 18 requirement-derived traces and six component ablations
Abstract:Dementia care increasingly involves connected sensors, medication devices, electronic records, and assistive technologies. Interoperability can transport observations but cannot maintain an accountable care state, reconcile evidence, determine who may act, or verify resolution. The shift from large language models to agentic engineering creates a systems opportunity: an external runtime can maintain memory across episodes, plan over goals and constraints, invoke tools, observe outcomes, and enforce governance. This paper presents Governed Closed-loop Agent Coordination (GCAC), an architecture for bounded agent participation in community dementia-care workflows. Evidence on care-coordination failures and policy obligations is translated into traceable system requirements. GCAC separates observation, governed memory, planning, deterministic policy enforcement, execution, and outcome monitoring through a typed event-memory-decision-action-outcome contract. A reference harness evaluates 18 evidence-derived traces covering missing records, medication conflict, caregiver reports, service failure, consent change, stale state, duplicate events, untrusted text, and suspected acute neurological change. GCAC satisfies all 18 contract oracles with zero policy-violating tool calls and correctly preserves obligations, rejects stale state, creates human hand-offs, and records workflow closure. Event-threshold and stateless-planner controls satisfy 2/18 and 1/18 oracles, respectively. Component ablations localise failures to the removed memory, policy, or versioning function. The results establish architectural conformance rather than clinical effectiveness and show how agentic systems can automate reconciliation, routing, documentation, and follow-up while preserving human authority over consequential care decisions.
[MA-10] When Does Execution Provenance Help Agent Memory Retrieval?
【速读】：该论文旨在解决语言智能体（language agent）在执行过程中，其历史记录超出上下文窗口（context window）限制时，记忆系统如何在严格令牌预算下准确检索完整支持性证据的问题。传统检索方法采用固定长度的令牌窗口和固定的k值评估指标，仅奖励单个片段的相关性，而无法判断整个证据集合是否能在有限上下文中容纳，导致证据碎片化或冗余信息引入。为克服这一局限，本文提出将代理-记忆检索建模为“带预算的证据补全”问题，并在共享源坐标中精确评分真实证据跨度（gold spans）。核心解决方案包括：首先基于工具调用参数与输出构建源对齐的溯源单元（source-aligned provenance units），进而引入零初始化残差图卷积网络（R-GCN）在类型化的溯源边上传播信息，以优化冻结的密集检索得分。实验表明，在1,207条独立保留的执行轨迹上进行2,000个基于跨度的记忆查询，相较于512令牌的扁平窗口，该方法在Full Support@2048指标上提升19.07点；即使相比各窗口大小下的最优单指标基准，仍高出11.96点。此外，在固定候选集与初始得分的前提下，图结构传播进一步带来4.55点的增益（95%置信区间[2.98, 6.18]），且该优势集中体现在真实证据跨越多个执行事件的情形。实体共现扩展未带来类似收益，关系与拓扑控制实验也证实该提升依赖于类型化变换及实际图结构。综上，源对齐候选单元有效缓解了粒度选择的主导权衡，而图条件传播则为分散证据提供了精准、局部的增强。
链接: https://arxiv.org/abs/2609.25913
作者: Yiqi Wang,Jinqian Ju,Jiaqi Zhang,Zequn Sun,Yiqun Duan,Mingkai Zheng,Taotao Cai
机构: University of Southern Queensland, Australia(南昆士兰大学); Southern University of Science and Technology, China(南方科技大学); Jiangsu University, China(江苏大学); Nanjing University, China(南京大学); Norve Labs Inc., Palo Alto, CA, USA(诺尔维实验室公司); University of Southern Queensland, Australia(南昆士兰大学)
类目: Multiagent Systems (cs.MA)
备注: 14 Pages, 4 Figures, 8 Tables
Abstract:A language agent’s execution history can exceed its context window, requiring its memory system to retrieve complete supporting evidence under a hard token budget. Evidence may span multiple execution events, yet conventional retrievers use fixed token windows and fixed-k metrics that reward individual fragments without showing whether the complete evidence set fits in context. Smaller windows reduce irrelevant text but scatter evidence across candidates, while flat-versus-graph comparisons can conflate candidate design with graph propagation. To address these limitations, we formulate agent-memory retrieval as budgeted evidence completion and score exact gold spans in shared source coordinates. We first construct source-aligned provenance units from tool arguments and outputs. We then apply a zero-initialized residual R-GCN to refine frozen dense-retrieval scores over typed provenance edges. We evaluate 2,000 span-grounded memory queries over 1,207 held-out execution-grounded ISETrace trajectories. With matched Dense-FT scoring, provenance units improve Full Support@2048 by 19.07 points over flat 512-token windows and remain 11.96 points above a per-metric oracle over four flat chunk sizes; the pattern also holds with cross-encoder scoring. Holding the candidates and seed scores fixed, graph propagation adds 4.55 points in Full Support@2048 (95% CI [2.98, 6.18]). This gain is concentrated when gold evidence spans multiple events; entity co-occurrence expansion produces no comparable benefit, and relation and topology controls confirm dependence on typed transformations and observed graph structure. Overall, source-aligned candidates address the dominant granularity trade-off, while graph-conditioned propagation adds a smaller, targeted benefit for distributed evidence.
[MA-11] Fully Byzantine-Resilient Multi-Agent Reinforcement Learning
【速读】：该论文旨在解决分布式拜占庭容错的演员-评论家多智能体强化学习（AC-MARL）中因通信层存在拜占庭攻击而导致智能体参数仅收敛至无攻击情形下极限点邻域的问题，从而造成性能下降。其核心解决方案是提出一种全容错的分布式方法——全容错AC-MARL（FRAC-MARL），该方法通过利用两跳通信信息中的冗余性来识别可信消息，实现对恶意通信干扰的有效抵御。在价值函数与团队奖励函数采用线性参数化、且拜占庭攻击局限于通信链路的前提下，证明了各智能体参数几乎必然收敛至与无攻击场景相同的极限点，即使在时变通信图下亦成立。研究进一步提出了一种新的拓扑条件以保证方法收敛性，并给出了可系统构建满足该条件网络的方案，同时证明该条件可在多项式时间内验证。实验部分在协作多机器人编队控制任务中验证了该方法的有效性。
链接: https://arxiv.org/abs/2609.25701
作者: Haejoon Lee,Dimitra Panagou
机构: University of Michigan(密歇根大学)
类目: Machine Learning (cs.LG); Multiagent Systems (cs.MA); Systems and Control (eess.SY)
备注: 12 pages, 3 figures
Abstract:We study distributed Byzantine-resilient actor-critic multi-agent reinforcement learning (AC-MARL), where agents collectively learn policies through local interactions. Existing methods guarantee convergence of the agents’ parameters only to a neighborhood of the attack-free limit points, resulting in degraded performance. We propose Fully Resilient AC-MARL (FRAC-MARL), a decentralized method in which each agent leverages redundancy in two-hop messages to identify reliable messages. Under linear parameterizations of the value and team-reward functions and Byzantine edge attacks, where adversarial behavior is confined to the communication layer, we prove that agents’ parameters converge almost surely to the same limit points as in the attack-free case over time-varying communication graphs. We introduce a novel topological condition for the convergence of our method, present a systematic method to construct such networks, and prove that this condition can be verified in polynomial time. Finally, we demonstrate our method on cooperative multi-robot formation control tasks.
[MA-12] pping Points in LLM -Based Multi-Agent Systems: Stance on Climate Change Action
【速读】：该论文旨在解决气候变化议题中公众意见动态演化机制的复杂性问题，特别是识别社会层面可能存在的“社会临界点”（social tipping points），即微小个体行为变化引发大规模集体态度转变的现象。其核心挑战在于如何在模拟环境中有效捕捉并分析此类非线性、涌现性的社会行为模式。解决方案的关键在于构建一种基于大语言模型（Large Language Models, LLMs）赋能的代理模型（Agent-Based Model, ABM），使模型中的个体代理不仅具备自然语言对话能力，还能通过叙事自我（narrative self）和情景记忆（episodic memory）维持连贯的个性特征与历史互动记录。研究通过双维度立场空间（信念强度与对制度信任度）量化代理间认知距离，并结合潜在狄利克雷分配（Latent Dirichlet Allocation, LDA）分析讨论主题演变，揭示了在微型社会系统中气候立场可能发生突变的现象。研究进一步指出，防止LLM固有偏见干扰对话动态、并保持代理人格与记忆的一致性，是确保模型可靠性与可解释性的关键方法论挑战。该框架为未来基于仿真推演设计真实世界应对气候变化等重大社会问题的干预策略提供了可行路径。
链接: https://arxiv.org/abs/2609.25432
作者: Astghik Altunyan,Shimon Edelman
机构: Cornell University(康奈尔大学)
类目: Multiagent Systems (cs.MA); Computers and Society (cs.CY)
备注:
Abstract:Because significant action to counter global warming requires massive public support, it is important to understand the dynamics of public opinion on climate issues. Of special interest are social tipping points, as revealed by large-scale effects of small perturbations in individual behaviors. Agent-based models (ABM) are an effective computational tool for studying these matters, because they allow controlled and systematic exploration of the effects of interventions that may be infeasible in real-world social systems. Large language models (LLMs) have been used to endow model agents with the ability to communicate in natural language (rather than by exchanging predefined messages), as well as with personality (in the form of a narrative self and episodic memory). We leverage LLM-powered ABM to look for tipping points in the social dynamics of a micro-society in which some of the discussions are about climate change. Our agents’ stance was defined by two variables: the strength of conviction about the urgency of climate action and the degree of trust in existing institutions. We quantified shifts in agents’ “beliefs” by monitoring, across multiple rounds of conversations, (1) inter-agent distances in this two-dimensional stance space and (2) the patterns of discussion topics as modeled by Latent Dirichlet Allocation (LDA). Our findings to date suggest that significant abrupt changes in climate-change stance do occur in this simple model. We report a number of methodological lessons from this study, notably, the need to prevent LLM biases from interfering with the conversational dynamics and, more generally, to maintain agent personality and episodic memories of interactions in the face of such biases. Resolving these issues may allow for using ABM-derived insights in designing real-life interventions vis-a-vis climate change and other important societal challenges.
[MA-13] Indirect tipping: a social attack surface in AI agent populations
【速读】：该论文旨在解决大规模部署生成式AI代理（Generative AI Agents）时所面临的系统性安全问题，即如何在群体层面保障其行为的稳定性与可控性。传统方法依赖“临界质量动力学”（Critical Mass Dynamics）来评估系统脆弱性，仅关注通过直接竞争颠覆协调均衡所需的最小攻击者比例，但该方法忽略了间接且更高效的操控路径。本文提出的关键解决方案是：构建一个能够刻画大规模语言模型（LLM）代理群体集体动态的分析框架，将不同协调均衡之间的关系映射为具有方向性和权重的拓扑结构，形成可导航的社会景观（social landscape）。研究发现，通过中间“踏脚石均衡”实现的间接突变（indirect tipping）可显著降低达成新状态所需的少数派投入，绕过对多数控制的要求，并开启原本无法通过直接挑战到达的转变路径。此外，可用替代状态的多样性及攻击时机的差异进一步重塑该景观，既带来控制机会，也引发非预期的不稳定性风险。因此，系统的抗干预能力并非固有属性，而是其与其它状态之间竞争关系的结构性特征。确保交互式AI代理群体的安全，必须结合个体代理能力、技术交互通道以及社会景观的整体测绘。
链接: https://arxiv.org/abs/2609.25194
作者: Ariel Flint,Luca Maria Aiello,Sara M. Constantino,Romualdo Pastor-Satorras,Andrea Baronchelli
机构: City St George’s, University of London (伦敦城市圣乔治学院，伦敦大学); IT University of Copenhagen (哥本哈根信息技术大学); Pioneer Centre for AI (先锋人工智能中心); Stanford University (斯坦福大学); Universitat Politècnica de Catalunya (加泰罗尼亚理工大学)
类目: Multiagent Systems (cs.MA); Artificial Intelligence (cs.AI); Computers and Society (cs.CY); Systems and Control (eess.SY); Physics and Society (physics.soc-ph)
备注:
Abstract:As generative AI agents are deployed at scale, safety will depend not only on technical safeguards and individual model design, but also on collective equilibria that determine how agent populations process information, prioritize actions, and respond to uncertainty. Yet the same equilibria that enable agents to coordinate also create a social attack surface. The standard framework to assess this vulnerability is critical mass dynamics: the minimum fraction of adversarial agents required to overturn an equilibrium through direct competition. Here, we show that this approach risks underestimating system vulnerability by reducing the problem to the identification of singular tipping points, and ignoring indirect but potentially more efficient routes through which collective behavior can be redirected. Through experiments with populations of LLM agents and an analytic framework that captures their collective dynamics at scale, we map critical-mass thresholds that define a directed, weighted topology over the space of coordination equilibria, and treat this topology as a navigable landscape. We show that indirect tipping through intermediate stepping-stone equilibria can reduce the committed minority required to reach an alternative state, bypass majority requirements, and make possible transitions inaccessible through direct challenges. The diversity of available alternatives and timing of the attack further reshape this landscape, creating opportunities for control as well as risks of unintended destabilization. These results show that an equilibrium’s resistance to committed intervention is not an intrinsic property but a structural feature of its competitive relations with alternative states. Securing populations of interacting AI agents therefore requires mapping this social landscape alongside individual agent capabilities and the technical channels through which they interact.
[MA-14] Qwen -Audio-Agent Technical Report
【速读】：该论文旨在解决语音交互系统中实时对话连续性与复杂任务执行效率之间的矛盾问题，即在保持全双工语音交互流畅性的同时，实现多步骤、异步任务的高效处理。其核心解决方案是提出一种基于前台-后台（foreground-background）架构的Qwen-Audio-Agent框架：前端代理（Frontend Agent）负责对话管理并决策是否直接调用工具或委派任务，后端代理（Backend Agent）在独立上下文中执行委托任务；由编排运行时（Orchestration Runtime）统一维护任务状态，协调用户输入与授权请求，并调度结果返回至对话流程。该运行时通过解耦语音中断、任务取消、执行完成与结果交付等关键事件，实现了对话与后台任务的并行推进。此外，环境事件和持久化记忆为跨会话提供上下文支持，且独立适配器设计增强了系统对不同前端模型、后端代理及客户端的可扩展性。实验表明，在134个真实驾驶舱场景测试中，混合执行模式（mixed execution）相较纯直接调用（72.39%）和全委派模式（80.60%）分别达到91.04%的任务成功率，并将平均任务执行延迟降低26.73%和30.91%，验证了直接工具调用与后台委派在不同任务类型中的互补优势。
链接: https://arxiv.org/abs/2609.25195
作者: Chong Deng,Yunjie Ji,Yuxiang Kong,Xiangang Li,Xu Li,Binbin Zhang,Haina Zhu,Jianheng Zhuo
机构: 未知
类目: Audio and Speech Processing (eess.AS); Multiagent Systems (cs.MA)
备注:
Abstract:We present Qwen-Audio-Agent, a harness that combines full-duplex voice interaction with asynchronous task execution through a foreground-background architecture. A Frontend Agent manages dialogue and selects between direct tool use and delegation, while a Backend Agent carries out delegated tasks in a separate context. An Orchestration Runtime maintains task state, coordinates requests for user input and authorization, and schedules the return of results to the conversation. The runtime separates speech interruption from task cancellation and execution completion from result delivery, allowing conversation to continue while delegated work proceeds. Environmental events and persistent memory provide context within and across sessions. Independent adapters support integration with different frontend models, backend agents, and clients. We instantiate the architecture in desktop assistance, intelligent cockpits, and voice customer service. On an in-house cockpit benchmark of 134 cases, mixed execution achieves a task success rate of 91.04%, compared with 72.39% and 80.60% for the direct and all delegated configurations, respectively. In a separate latency evaluation on matched successful turns, mixed execution reduces mean task execution latency by 26.73% and 30.91% relative to these baselines, respectively. These results support the complementary use of direct tool calls for immediate operations and backend delegation for multi-step tasks.
自然语言处理
[NLP-0] Flash-dLLM : IO-Aware KV Caching and Parallel Decoding for Fast Memory-Efficient Diffusion LLM s
【速读】： 该论文旨在解决生成式 AI（Generative AI）中扩散型大语言模型（Diffusion Large Language Models, dLLMs）在实际部署中面临的推理效率低下问题，其核心挑战在于缺乏高效的键值（Key-Value, KV）缓存机制与可扩展的并行解码方法。现有加速方案通常孤立地优化KV缓存或并行解码，忽略了在联合使用缓存复用与并行令牌验证时所引发的显存输入/输出（I/O）瓶颈。本文提出了一种无需训练的推理加速框架——Flash-dLLM，其关键创新在于：首先识别出GPU显存I/O是启用KV缓存的dLLM推理中的主要性能瓶颈，并通过设计一种I/O感知的融合式KV缓存核函数，显著减少冗余内存访问；在此优化的缓存机制基础上，进一步提出一种基于KV缓存驱动的“草稿-验证”解码策略，使dLLM自身同时担任草稿生成与验证角色，无需依赖额外的辅助模型。该统一架构在保持生成质量的同时，实现了更高的解码速度、更强的长序列与大规模批量处理能力。大量实验表明，Flash-dLLM在数学推理与代码生成基准测试中均显著优于现有最优方法，在GSM8K和HumanEval上分别实现5.1倍和11.0倍于最强基线Elastic-Cache的加速比，兼具卓越的推理速度与内存效率。
链接: https://arxiv.org/abs/2609.26796
作者: Quan Nguyen-Tri,Mukul Ranjan,Zhiqiang Shen
机构: VILA Lab, MBZUAI(阿布扎比人工智能大学视觉智能实验室); Abu Dhabi, UAE(阿布扎比, 阿联酋)
类目: Computation and Language (cs.CL)
备注: Code available at: this https URL
Abstract:Diffusion Large Language Models (dLLMs) have recently emerged as a promising alternative to autoregressive LLMs by enabling non-autoregressive text generation. However, their practical deployment remains limited by inefficient inference, largely due to the absence of effective Key-Value (KV) caching and scalable parallel decoding mechanisms. Existing acceleration methods typically study KV caching and parallel decoding in isolation, overlooking the I/O bottlenecks that arise when cache reuse and parallel token verification are jointly applied. In this work, we introduce \textbfFlash-dLLM , a training-free inference acceleration framework for fast and memory-efficient dLLMs. Flash-dLLM first identifies GPU memory I/O as a dominant bottleneck in KV-cache-enabled dLLM inference and addresses it with an I/O-aware fused KV-cache kernel that reduces redundant memory movement. Building on this optimized cache mechanism, Flash-dLLM further proposes an efficient KV-cache-driven draft-and-verify decoding strategy, where the dLLM itself serves as both drafter and verifier without requiring an auxiliary model. This unified design enables faster decoding while preserving generation quality and improving scalability to longer sequences and larger batch size. Extensive experiments on mathematical reasoning and code-generation benchmarks demonstrate that Flash-dLLM consistently outperforms existing state-of-the-art dLLM acceleration methods in both inference speed and memory efficiency. In particular, it achieves 5.1\times and 11.0\times speedups over prior strongest baseline Elastic-Cache on GSM8K and HumanEval, respectively.
[NLP-1] Beyond Repeated Sampling: Learning Search Policies for LLM Reasoning
【速读】： 该论文旨在解决大语言模型在处理复杂推理任务时，依赖大量测试阶段计算但效率低下的问题。当前主流的“朴素重复采样”策略通过独立生成多个答案尝试来提高正确率，但由于仅依赖局部解码噪声进行探索，导致生成结果高度重复，缺乏语义层面的多样性。其核心问题是：如何实现更高效、更具探索性的推理路径搜索？论文提出的关键解决方案是引入语义级探索机制，即在生成答案前先主动采样与问题相关的概念、提示或策略，并以此作为条件引导答案生成。进一步地，作者设计了一种可训练的概念生成器，利用强化学习优化其输出，使其生成的概念能够最大化下游大型冻结答案生成器的成功率（pass@k）。实验表明，该方法在数学推理等难题上显著优于传统重复采样，不仅超越了使用更大但未调优模型所提取的概念，还具备跨模型家族的泛化能力。因此，一个小型概念生成器可被训练为一个高效且可复用的搜索策略，用于指导更大规模模型的推理过程。
链接: https://arxiv.org/abs/2609.26704
作者: Ismail Labiad,Matthieu Kowalski,Marc Schoenauer,Rémi Munos,Julia Kempe
机构: Meta FAIR; Université Paris-Saclay, LISN, Inria, CNRS; NYU Courant Institute and CDS
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
备注:
Abstract:Large language models increasingly tackle hard reasoning problems by spending more test-time compute, yet the dominant strategy remains naive repeated sampling: draw many independent solutions and hope one is correct. Because such sampling explores only through local decoding noise, it tends to produce many near duplicate attempts rather than genuinely different ideas. We ask whether exploration can instead be steered at a semantic level, by first sampling problem specific concepts, hints, or strategies and then conditioning answer generation on them. We refine this into a simple, more exploratory procedure that emits many diverse concepts in a single trajectory, and evaluate it on hard problems where repeated sampling struggles. We then go a step further and make concept generation trainable: a small concept generator is optimized with reinforcement learning so that its concepts maximize the downstream success of a larger, frozen answer generator. On hard mathematical reasoning problems, the trained concept generator substantially improves the answer generator’s pass@k over naive repeated sampling at the same answer generation allocation, surpasses concepts drawn from much larger untuned models, and transfers to answer generators it was never trained against, including a model from a different family. A small model can thus be trained into an effective, reusable search policy for a much larger one.
[NLP-2] Measuring the Serving Stack Instead of the Model: Hidden Confounds in Local Tool-Use Evaluation EMNLP2026
【速读】： 该论文旨在解决生成式 AI（Generative AI）在实际部署中因本地服务栈（local serving stacks）对工具调用（tool call）协议处理差异而导致评估结果失真的问题。其核心挑战在于，当前评估框架通常假设模型行为是决定工具调用成功率的唯一因素，而忽略了服务层（如Ollama、vLLM、SGLang等）在解析和传递工具调用时的底层实现差异。解决方案的关键在于：将服务栈的行为纳入评估协议的考量范围，并提出一套系统性检查清单，以确保评估结果真实反映模型能力而非受服务层配置或处理逻辑干扰。具体而言，研究发现不同服务栈对同一请求的处理存在显著差异——例如，某些模型在Ollama中被静态模板标志拒绝，导致无法执行推理；而未正确保留结构化失败元数据会使得这些拒绝被误判为“模型未产生工具调用”，进而导致精度（fidelity）评估出现严重偏差。通过引入文本工具列表并保留原生工具调用通道，可恢复多数已接受模型的测量保真度；但统一使用文本协议反而降低了具备原生工具调用支持的Llama-3.2模型的表现。此外，约束解码虽能减少解析失败，但可能引发非终止问题，且实例级与池化级估计结果差异可达约55个百分点。因此，该研究强调必须将服务层行为作为评估协议的一部分进行标准化管理。
链接: https://arxiv.org/abs/2609.26693
作者: Lijuan Tang,Yuemeng Zheng
机构: Northeastern University, Seattle(东北大学西雅图分校)
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI); Software Engineering (cs.SE)
备注: 9 pages, 4 figures, 3 tables. Accepted at the 2nd Workshop for Research on Agent Language Models (REALM) @ EMNLP 2026
Abstract:A coding agent must emit a valid tool call–a parseable invocation of a tool in the provided schema–before the harness can execute its chosen action. We study how local serving stacks affect this protocol step and show that measured outcomes can depend on the serving layer rather than model behavior alone. In Ollama, the default tools= request is gated per model by a static template flag: some models are accepted and return calls as text, some return native tool_calls, while Phi-3 and Gemma-3 are rejected before inference. In our harness, rejection and retry exhaustion are not preserved as structured failure metadata, so downstream analysis can misclassify them as model non-calls and naively report 0% fidelity. Adding a text tool list while retaining the native channel recovers much of the measured fidelity for accepted models, whereas a uniform text protocol reduces fidelity for Llama-3.2, which has native tool-call support. Cross-stack probes on Ollama, this http URL, vLLM, and SGLang show different handling of the same request. Constrained decoding removes parse failures but can induce non-termination, and turn-pooled versus per-instance estimates differ by up to about 55 points. We conclude with a checklist for treating serving behavior as part of the evaluation protocol.
[NLP-3] Detecting GPT -Assisted Writing Using Interpretable Stylometric Features
【速读】： 该论文旨在解决学术界日益严峻的难题：如何有效区分学生独立撰写的文本与借助生成式 AI（Generative AI）辅助完成的文本。其核心解决方案在于利用仅从提交文本中提取的可解释性风格特征（interpretable stylometric features），通过机器学习方法实现对 GPT 辅助写作的判别。研究基于90名参与者分别独立撰写和使用 ChatGPT 辅助撰写的文本数据，采用交叉验证策略保持同一参与者的数据不被拆分，评估了八种分类器性能。结果显示，随机森林（Random Forest）在留出测试集上达到 0.87 的 ROC-AUC 和 0.84 的 F1 分数，误报率（False Positive Rate）为 22.2%，漏报率（False Negative Rate）为 11.1%。SHAP 分析进一步揭示，词汇与语法特征是驱动模型预测的关键因素。研究证明，基于文本内在属性的透明化特征能够提供可量化的信号，用于检测 GPT 辅助写作行为。
链接: https://arxiv.org/abs/2609.26687
作者: Rajesh Kumar,Nabeel Siddiqui,Alexander Fuchsberger
机构: Bucknell University (巴克内尔大学); Susquehanna University (萨斯奎汉纳大学)
类目: Computation and Language (cs.CL); Computers and Society (cs.CY)
备注: 10 pages, 6 figures, 5 tables
Abstract:Distinguishing GPT-assisted from independently authored student writing has become a critical challenge in academia. This paper evaluates the discriminative capability of interpretable stylometric features extracted solely from submitted text. Using data from 90 participants who wrote both independently and with ChatGPT assistance, we evaluate eight machine learning classifiers while keeping data from the same participant together during validation. On the held-out test set, Random Forest achieved an ROC-AUC of 0.87 and an F1-score of 0.84, with False Positive and False Negative rates of 22.2% and 11.1%, respectively. SHAP analysis shows that lexical and grammatical characteristics drive the resulting predictions. The findings suggest that transparent, text-intrinsic features provide measurable signal for detecting GPT-assisted writing.
[NLP-4] Diffusion Drafts AR Verifies: Accelerating Document OCR with Self-Speculative Decoding
【速读】： 该论文旨在解决自回归光学字符识别（OCR）视觉-语言模型在推理速度上的瓶颈问题，即传统方法需逐个解码输出标记（token），导致效率低下。其核心挑战在于，尽管基于扩散模型的并行生成在理论上可加速处理，但因多个标记在单次扩散步骤中被并行预测且彼此依赖关系未明确，直接提交结果易引入错误。为此，本文提出GravityOCR，一种参数共享的自回归块-扩散混合模型，通过联合训练并行草稿生成与因果自回归验证路径来解决此问题。其关键创新在于：利用因果自回归路径对并行草稿进行验证后再决定是否提交，从而实现在不引入独立草稿网络的前提下，每轮前向传播可提交多个输出标记；同时，该结构支持基于序列级和结构级OCR奖励的广义强化学习策略优化（GRPO），避免了对扩散轨迹似然性的估计，直接优化共享草稿器参数。实验表明，在OmniDocBench v1.6上，该方法将整体得分从94.92提升至95.16，且保持了高效的扩散草稿能力，最终模型性能接近原始GLM-OCR的95.48分；在SGLang部署环境下，GravityOCR平均每轮提交9.7个标记，实现了3.94倍的仅解码速度提升及1.32倍的端到端页面处理速度提升。
链接: https://arxiv.org/abs/2609.26638
作者: Dohyun Kim,Sungjun Han,Hyungguk Kim,Yusik Kim,Jamin Shin,Paul Hongsuck Seo,Hongjoon Ahn
机构: Trillion Labs; Korea University(韩国大学); Seoul National University(首尔国立大学)
类目: Computation and Language (cs.CL); Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Autoregressive OCR vision-language models accurately convert document images into text and structured markup, but require one sequential decoding step per output token, limiting inference speed. Unlike open-ended text generation, OCR outputs are strongly grounded in the input image, making diffusion-based parallel generation promising. However, when several tokens are predicted in one diffusion step, each is predicted before the others are known. Committing them directly can therefore introduce errors. We therefore introduce GravityOCR, a parameter-shared AR-block-diffusion model jointly trained for parallel drafting and causal AR verification. Verifying drafts before commitment lets the model commit multiple output tokens per round without a separate drafting network. The causal AR path also enables GRPO with sequence- and structure-level OCR rewards, avoiding diffusion-trajectory likelihood estimation while updating the shared drafter parameters. On OmniDocBench v1.6, AR-path GRPO improves the Overall score from 94.92 to 95.16 without reducing diffusion drafting efficiency, while the final model remains close to the original GLM-OCR score of 95.48. In an SGLang serving deployment, GravityOCR commits an average of 9.7 output tokens per forward pass and achieves a 3.94\times decode-only speedup on region crops and a 1.32\times end-to-end page-processing speedup over AR decoding.
[NLP-5] Capable yet Parsimonious: Extracting and Characterizing Hidden Chain-of-Thought in Frontier Models
【速读】： 该论文旨在解决当前前沿语言模型（frontier language models）中推理能力评估的可验证性问题，即由于闭源系统无法获取原始思维链（Chain-of-Thought, CoT）轨迹，导致其推理过程难以被真实观测与分析。为克服这一限制，论文提出通过标准API接口注册自定义工具，诱导前沿模型将中间推理过程外化为可观测的推理痕迹。其解决方案的关键在于：设计一种可扩展的外化机制，在不依赖模型内部结构的前提下，使闭源模型能够输出可分析的推理路径，并通过在开源模型上进行对比验证，确认所提取的推理轨迹与原生CoT具有相当性能表现，显著优于无推理基线。进一步地，研究系统性地分析了前沿模型在推理过程中的结构特征，包括推理步骤的类型分布、生成效率以及推理树的构建方式，发现如GPT-6 Astra这类模型展现出高效率的定向推理模式——早期即选择正确路径，仅将关键推理步骤外部化，其余基础计算则在内部完成。该方法为理解前沿模型的内在推理行为提供了超越基准测试分数的行为视角。
链接: https://arxiv.org/abs/2609.26637
作者: Xiaoyu Luo,Tao Ren,Wenrui Yu,Xiao Li,Qiongxiu Li,Johannes Bjerva
机构: Aalborg University (奥尔堡大学); Seafill
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI); Cryptography and Security (cs.CR)
备注: 33 pages,14 figures
Abstract:The rapid capability gains of frontier language models are widely attributed to improved reasoning abilities, yet this cannot be verified as raw CoT traces in closed-source systems are hidden. By registering a simple custom tool through a standard API feature, we induce frontier models to externalize intermediate reasoning. Because these traces may reflect post-hoc rationalization rather than genuine reasoning, we first evaluate against native CoT on open-source models and extend to closed-source frontier models including GPT-6 Astra. We find that the extracted reasoning matches native reasoning performance and substantially outperforms no-reasoning baselines, across competition mathematics, science, and code generation. We then characterize how frontier models structure their intermediate reasoning. Across token efficiency, reasoning-step types, and induced reasoning trees, we identify systematic differences in how models externalize, compress, and organize reasoning. We find that Astra exhibits token-efficient directed reasoning, selecting a correct trajectory earlier, while resolving elementary steps internally and externalizing only crucial reasoning. These findings provide a behavioral lens on frontier-model reasoning beyond benchmark scores.
[NLP-6] Knowledge Pull Requests for Continual Document Authoring
【速读】： 该论文旨在解决持续性文档撰写中知识更新与可解释性之间的矛盾问题，即在面对来自新源、多语言或不同时期的新增知识时，现有方法要么缺乏对知识变更内容的追踪（如直接修改但不记录变更依据），要么采用从头重写的方式，导致信息冗余且难以追溯变化来源。其解决方案的核心是提出“知识拉取请求”（Knowledge Pull Requests, KPRs）框架，通过结构化流程实现可解释的知识集成：首先提取新增知识中的主张（claim），经筛选与路由后分配至文档相应章节，并自动检测与原文内容的潜在冲突；最终生成包含“变更日志”（ChangeLog）的可追溯记录，将知识层面的变化（主张提案）与文本层面的修改（文档差异）明确分离。实验表明，相较于直接重写或从头生成，KPRs在多语言维基百科修订及基于检索增强生成（RAGTIME）的查询驱动报告更新任务中，能够更高效地整合信息、更好保留原有内容，并在单位生成词元内传递最多有效知识；此外，经KPR修订的文章在问答任务中表现优于依赖搜索的前沿模型，尤其在处理仅存在于其他语言中的知识时具备更强的覆盖能力。
链接: https://arxiv.org/abs/2609.26634
作者: Alexander Martin,Benjamin Van Durme
机构: Johns Hopkins University (约翰霍普金斯大学)
类目: Computation and Language (cs.CL)
备注: Code: this https URL
Abstract:We introduce Knowledge Pull Requests (KPRs), a framework for continual document authoring that makes each change interpretable. Documents require ongoing revision as new knowledge surfaces from other sources, languages, or times, but existing approaches either edit with no account of what knowledge changed or regenerate from scratch. A KPR integrates new knowledge into a document by extracting claims, filtering and routing them to sections, and flagging conflicts with existing content, producing a ChangeLog that separates what knowledge changes (claim proposal) from how the text changes (document diff). We evaluate KPRs on revising Wikipedia across languages and updating query-driven reports on RAGTIME. KPRs integrate more information and better preserve existing content than rewriting from sources or regenerating from scratch, while adding the most information per token generated. A KPR-revised article also grounds question answering better than a frontier model with search, which does not surface knowledge documented only in other languages.
[NLP-7] PERSONAWEAVER: Controllable Diversity Beyond Conventional Archetypes in Procedural Character Generation
【速读】： 该论文旨在解决生成式角色在游戏、仿真等虚拟世界中因依赖大语言模型（LLM）而产生的行为同质化问题。现有方法要么直接由LLM生成角色，要么从角色库中调用预设人格模板，均导致生成的角色在道德立场和交互行为上高度趋同，普遍表现出符合正向道德规范且类似助手的回应模式。为突破这一局限，论文提出PersonaWeaver，其核心创新在于将世界构建（world building）与行为设定（behavioral specification）解耦，并通过人工精心设计的多元道德立场与对话反应库来建模角色行为。该方案有效引导LLM脱离默认行为模式，在十种现实与奇幻场景下，显著拓宽了道德取向与互动反应的分布范围，提升了人际语言多样性、响应长度及情感倾向的丰富性，同时减少了典型化属性组合的出现。该方法的关键在于通过可控制的、多样化的外部知识注入，实现对角色行为的精准差异化调控。
链接: https://arxiv.org/abs/2609.26629
作者: Maan Qraitem,Kate Saenko,Bryan A. Plummer
机构: Boston University (波士顿大学)
类目: Computation and Language (cs.CL)
备注: Accepted at the 1st PANDORA Workshop: Pluralistic AI and NLP
Abstract:Procedural character generation aims to populate games, simulations, and other virtual worlds with diverse characters. Large language models (LLMs) offer a promising foundation for scaling this task. However, LLM-based procedural character generation remains at an early stage: existing methods either generate characters directly or adapt profiles retrieved from persona banks. As we show, both approaches produce behaviorally homogeneous populations: characters overwhelmingly agree with positive moral norms and respond to questions with helpful, assistant-like reactions. To mitigate this homogenization, we introduce PersonaWeaver, which disentangles world building from behavioral specification and models behavior through setting general, diverse, manually curated banks of moral positions and conversational reactions. This design allows us to test how far LLM(s) can be pushed beyond their default behavioral patterns across settings. Across ten realistic and fantastical settings and three LLM(s), PersonaWeaver produces broader moral and interactional response distributions than prior work. Its guidance also diversifies interpersonal language, response length, and sentiment. It also produces less archetypal combinations of world attributes. Code is available at this https URL.
[NLP-8] Semantic Abstraction for Natural Language Inference: a Methodological Framework for Discovering and Compensating Semantic Knowledge and Reasoning Gaps in Large Language Models
【速读】： 该论文旨在解决大语言模型（LLM）在自然语言推理（NLI）任务中面临的语义抽象能力不足问题，尤其体现在对隐含意义、上下文概念关系及词句间语义关联的捕捉能力有限。其核心挑战在于模型缺乏对抽象语义知识的有效利用，导致在非蕴含类（non-entailment）任务中表现不佳。解决方案的关键在于提出一种方法论框架，通过重构前提与假设之间的词汇-语义关系，基于语义相容性（semantic compatibility）与不相容性（semantic incompatibility）的概念，在更高层次上构建新的抽象语义知识。该框架重新配置语义网络，引入更灵活的推理路径，使模型在不同情境下产生一致且可共识的响应。实验表明，该方法能有效发现并弥补模型在语义知识上的空白，显著提升准确率，尤其在非蕴含类别上提升超过10%。研究强调，仅增加数据量不足以填补推理鸿沟，关键在于提供结构化的语义支撑；该混合方法聚焦于被忽视的词语间关系，促使模型合成缺失信息。未来方向不应局限于扩大模型规模，而应构建类人思维般灵活的语义骨架（semantic scaffolding），以推动具备更强鲁棒性和可解释性的智能体发展，实现更可靠的自然语言理解。
链接: https://arxiv.org/abs/2609.26610
作者: David Torres-Moreno,Jorge Hermosillo-Valadez
机构: Centro de Investigación en Ciencias, Universidad Autónoma del Estado de Morelos(莫雷洛斯自治大学科学研究中心); CONACYT MEXICO(墨西哥国家科学与技术委员会)
类目: Computation and Language (cs.CL)
备注: 59 pages, 13 figures. Preprint of the article published in Knowledge-Based Systems, this https URL
Abstract:Despite their outstanding performance on many NLP tasks, LLMs face serious challenges related to semantic abstraction. In this study, we are interested in understanding how LLMs leverage abstract semantic knowledge in natural language inference (NLI), which requires sophisticated linguistic capabilities to interpret implicit meanings, contextual conceptual relationships, and semantic connections between words and phrases. To this end, we propose a methodological framework for constructing new semantic knowledge at a higher level of abstraction, which we define under the notions of semantic compatibility and incompatibility for NLI. In this framework, the meaning of the lexical-semantic relations between the premise and the hypothesis is reconfigured to achieve a more flexible semantic network that induces different reasoning paths in LLMs. These new pathways show a consistent pattern of responses that allows agreement on a single response. The results demonstrate that our proposal allows to discover and compensate for LLMs’ semantic knowledge gaps in NLI, achieving significant improvements in accuracy, exceeding 10% for some models, and in particular for the non-entailment class. It is essential to note that LLMs need structured knowledge and not just more data to bridge reasoning gaps. Our hybrid approach directs attention to overlooked word relationships, allowing models to synthesize missing information. We believe that the future lies not in increasing model size, but in creating a semantic scafolding that mimics the flexibility of human thinking. Hopefully, our proposal will enable the development of more robust agents and interpretable reasoning, guiding AI toward reliable language understanding.
[NLP-9] A retrospective analysis on the use of LLM s to study infant syntax learning
【速读】： 该论文旨在解决当前大型语言模型（Large Language Models, LLMs）在模拟婴幼儿早期句法习得过程中的方法论局限性与理论适切性问题。其核心关切在于，尽管“BabyLM”挑战等研究项目致力于构建能够达到人类水平句法表现、且基于发展现实语料（developmentally realistic corpora）训练的模型，但现有研究在数据构建、模型选择、训练策略及句法评估等方面存在显著的方法论假设，这些假设限制了研究结论的理论普适性。论文的关键发现在于，即便采用发展现实语料进行训练，大型语言模型在主流句法基准测试上的表现提升有限，这暗示了大型语言模型与真实婴儿句法学习者之间存在重要的计算机制差异，表明当前基于神经网络的生成式AI（Generative AI）在模拟儿童语言发展方面仍难以充分捕捉认知层面的本质特征。
链接: https://arxiv.org/abs/2609.26539
作者: Hélie Bazin(SCAI, SND, ISIR),Anouk Barberousse(SND),François Yvon(MLIA)
机构: 未知
类目: Computation and Language (cs.CL)
备注:
Abstract:Large language models (LLMs) have increasingly been used to investigate how children acquire syntax at an early stage of development. This is notably the central scientific goal of the BabyLM challenge, a community-wide effort to develop models that achieve human-level syntactic performance while being trained on developmentally realistic corpora. In this paper, we reflect on the use of LLMs in the study of infant syntax learning by providing an epistemological assessment of several studies from this research program. We discuss how datasets are built, which models are implemented, how they are trained and syntactically evaluated. We observe significant assumptions in the methodology of BabyLM and related studies, thus mitigating their theoretical scope. We additionally observe that using developmentally-realistic corpora have limited effects on models performance on commonly-used benchmarks, which suggest important computational differences between LLMs and the infant syntax learner.
[NLP-10] ranscribe Translate and Optimize: Joint Reward Learning for Speech Translation
【速读】： 该论文旨在解决大语言模型（LLM）驱动的语音翻译中，基于转录文本的思维链（Chain-of-Thought, CoT）方法在监督微调（SFT）阶段使用参考转录文本，而在推理阶段采用模型自动生成的转录文本所导致的训练-推理不匹配问题。其核心解决方案是通过分组相对策略优化（Group Relative Policy Optimization, GRPO）实现语音识别与翻译的联合微调，将翻译结果建模为依赖于模型生成转录文本的条件输出，并对比三种不同的词元优势（token advantage）策略。实验基于Qwen2.5-Omni-3B模型，在四种语言上对CoT与直接语音翻译（Direct Speech Translation, Direct ST）在SFT和GRPO两种微调方式下进行评估，分别在CoVoST 2和FLEURS数据集上测试。结果表明，CoT GRPO相较于Direct ST GRPO分别在CoVoST 2和FLEURS上提升1.77和0.83个平均BLEU分数；相较于CoT SFT，GRPO进一步提升0.82和0.67个BLEU分数，并相对降低8.8%和7.2%的词错误率（Word Error Rate, WER）。这表明强化学习微调能够有效缓解训练-推理不一致问题，同时协同提升语音识别与机器翻译性能。
链接: https://arxiv.org/abs/2609.26536
作者: Yanghe Dong,Wanting Huang,Weiran Wang
机构: 未知
类目: Computation and Language (cs.CL); Audio and Speech Processing (eess.AS)
备注: 5 pages
Abstract:In LLM-based speech translation, transcription-based chain-of-thought (CoT) suffers from a mismatch between reference transcripts used in supervised fine-tuning (SFT) and model-generated transcripts at inference. To address this, we propose joint recognition and translation fine-tuning via group relative policy optimization (GRPO). We score both transcripts and translations, with translation conditioned on model-generated transcripts, and compare three token advantage strategies. Using Qwen2.5-Omni-3B across four languages, we evaluate CoT against direct speech translation (Direct ST) under SFT and GRPO, training on CoVoST 2 and testing on CoVoST 2 and FLEURS. CoT GRPO outperforms Direct ST GRPO by 1.77 and 0.83 average BLEU points on CoVoST 2 and FLEURS. Compared to CoT SFT, GRPO boosts BLEU by 0.82 and 0.67 points and reduces word error rate (WER) by 8.8% and 7.2% relatively. These results highlight reinforcement fine-tuning as an effective method to mitigate the training-inference mismatch, jointly improving recognition and translation.
[NLP-11] A Semiotics-Aware Framework for Evaluating Fidelity and Coverag e in Natural Language Generation
【速读】： 该论文旨在解决传统文本相似性度量方法在评估两段描述相同表达的文本时，难以捕捉其语义框架差异的问题。标准方法依赖词汇重叠或整体文本相似度，无法有效反映文本在语境意义和话语指涉上的细微差别。为此，论文提出一种评估文本间符号学对齐（semiotic alignment）的框架，其中符号学特征（semiotic profile）涵盖文本所凸显的上下文含义及话语引用。该方法生成两个核心指标：符号学保真度（Semiotic Fidelity）与符号学覆盖度（Semiotic Coverage），分别衡量一个文本的符号学特征在另一文本中得到支持的程度，以及前者对后者符号学特征的还原能力。实验结果表明，覆盖度通常低于保真度，且大语言模型（LLM）与人工标注数据之间的对齐程度在低采样温度下最高，而较高温度会削弱这种对齐。其解决方案的关键在于引入符号学视角，通过双维度指标实现对文本语义框架差异的量化分析。
链接: https://arxiv.org/abs/2609.26527
作者: Lorenzo Zangari,Davide Picca
机构: University of Lausanne(洛桑大学); Lausanne, Switzerland(洛桑，瑞士)
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
备注:
Abstract:When two texts describe the same expression, standard metrics based on lexical overlap or whole-text similarity may fail to detect meaningful differences in how that expression is framed. We propose a framework to evaluate semiotic alignment between texts, where a semiotic profile encompasses both the contextual meaning and the discourse references made salient by a text. Our approach yields two scores, Semiotic Fidelity and Semiotic Coverage, estimating how much of one text’s profile is supported by the other and how much of the other’s profile it recovers. Experiments show that coverage is typically lower than fidelity, and that alignment between LLMs and human-curated data is highest at low sampling temperatures, while higher temperatures reduce this alignment.
[NLP-12] Calibration as a First-Class Criterion in LLM Evaluation EMNLP2026
【速读】： 该论文旨在解决大语言模型（Large Language Models, LLMs）在实际应用与研究中普遍存在的置信度校准不足问题，即模型所表达或隐含的置信度与其实际正确性之间缺乏一致性。尽管已有测量校准性的方法，但其在主流自然语言处理（Natural Language Processing, NLP）研究中的采纳率极低——新模型、数据集和基准测试往往不验证其置信度是否可靠。这一采纳鸿沟已成为可信LLM评估的主要障碍。校准失准在两个关键领域引发严重后果：部署阶段，过度自信的错误可能导致实质性损害；研究流程中，诸如“以大模型为裁判”（LLM-as-a-judge）、合成数据生成及主动学习等依赖于可靠置信度的方法，若未验证校准性，则可能引入系统性偏差。现有标准校准指标仅需每样本的置信度分数与正确性判断，而当前多数基准已具备这两项输入，因此校准性能可立即报告。然而，对于开放生成任务，如何定义这两个核心输入仍是一个未解挑战。为此，论文主张各NLP子领域应将主要性能指标与校准得分配对使用，并呼吁将校准性视为模型的基本属性，而非边缘化议题。
链接: https://arxiv.org/abs/2609.26489
作者: Mario Sanz-Guerrero,Katharina von der Wense
机构: Johannes Gutenberg University Mainz(美因茨约翰内斯古滕贝格大学), Germany; University of Colorado Boulder(科罗拉多大学博尔德分校), USA
类目: Computation and Language (cs.CL)
备注: Accepted to the 3rd Workshop on Uncertainty-Aware NLP (UncertaiNLP) at EMNLP 2026
Abstract:Calibration of language models – the alignment between expressed or implicit confidence and empirical correctness – is a well-studied subfield within NLP. Methods to measure it already exist. The problem is adoption: outside this subfield, NLP research regularly introduces new models, datasets, and benchmarks without checking whether the model’s confidence scores are meaningful. We argue that this adoption gap is a major obstacle to trustworthy LLM evaluation. Miscalibration causes problems in two distinct areas: at deployment, where overconfident mistakes cause real harm, and inside the research pipeline, where methods like LLM-as-a-judge, synthetic data generation, and active learning rely on calibrated confidence without verifying it. Standard calibration metrics only require two inputs per example: a confidence score and a correctness judgment. Most benchmarks in use today already provide both, meaning calibration can be reported immediately. For open-ended generation, however, defining these two inputs is still an open challenge. We argue that each NLP subfield should pair its main performance metric with a calibration score and call for treating calibration as an essential property of every model rather than a niche topic.
[NLP-13] Spoken Language Models that Think Aloud
【速读】： 该论文旨在解决生成式语言模型在口语化场景中应用时，因采用串行“思考-说话”（think-then-speak）范式而导致的长段沉默间隔问题，这会严重影响实时语音交互的流畅性。其核心解决方案是提出一种异步“自言自语”（asynchronous think-aloud）框架，嵌入到思考者-说话者（Thinker-Talker）架构中，通过并行维护一个主推理流（用于逻辑推导）与一个轻量级的“自言自语”流（生成基于用户输入和推理状态的任务相关进展话语），实现推理过程中的持续语音反馈。关键在于引入动态平衡策略，在运行时协调两个流：当检测到可能产生沉默间隙时触发额外的自言自语输出，而在最终回答就绪时取消待发送的中间语音，从而在保持与串行基线相当的答案准确率的前提下，显著减少用户可感知的沉默时间，有效提升口语化语言模型（Spoken Language Models, SLMs）的响应性与交互自然度。
链接: https://arxiv.org/abs/2609.26488
作者: Junyi Ao,Kainan Peng,Mingbo Ma,Shun Zhang,Zhenyu Tang,Xutai Ma,Xiang Li,Yinghao Li,Yuancheng Wang,Zhizheng Wu,Haizhou Li,Qing He,Xubo Liu
机构: Meta Superintelligence Labs; The Chinese University of Hong Kong, Shenzhen
类目: Computation and Language (cs.CL); Sound (cs.SD); Audio and Speech Processing (eess.AS)
备注: Accepted at SLT 2026
Abstract:While Chain-of-Thought (CoT) reasoning has improved the capability of language models, directly applying it to Spoken Language Models (SLMs) may introduce long silent intervals under the serial “think-then-speak” paradigm, disrupting real-time spoken interaction. To address this issue, we propose an asynchronous think-aloud framework for reasoning-based SLMs within the Thinker-Talker architecture. The framework maintains a primary reasoning stream for logical deduction and a lightweight think-aloud stream that generates short, task-grounded progress utterances conditioned on the user input and the evolving reasoning state. A dynamic balance strategy coordinates the two streams at runtime, triggering additional think-aloud speech to avoid silent gaps and canceling pending utterances when the final response becomes ready. Experiments on spoken reasoning and question-answering benchmarks show that our approach substantially reduces user-audible silence during reasoning while maintaining answer accuracy comparable to that of a serial “think-then-speak” baseline, demonstrating the potential of asynchronous think-aloud for responsive interaction in SLMs.
[NLP-14] How to Estimate Whether You Have Found Several Needles in a Haystack: Measuring Calibration in Multi-Label Text Classification
【速读】： 该论文旨在解决多标签分类（multi-label classification）中置信度校准（confidence calibration）的可靠性问题，尤其针对标签数量庞大且负样本占主导的场景（如临床笔记的医学编码或新闻主题识别）。现有校准评估方法在多标签任务中普遍存在偏差：传统的分箱（binning）策略要么低估校准误差，要么仅反映标签出现频率，或因大量空箱导致统计不稳健。其核心解决方案是提出一种新型分箱方案，通过为正样本和负样本赋予相等权重，实现标签层面（label-wise）的均衡校准误差估计。实验表明，该方法在层次化与极端多标签分类任务中均能提供有意义的校准误差评估，显著优于现有方法。同时，研究揭示了对大语言模型（large language models）进行多标签预测置信度校准仍是一个开放性挑战，为后续研究奠定了可靠的评估基准。
链接: https://arxiv.org/abs/2609.26468
作者: Sophie Henning,Georg Hofmann,Alexander Schulte,Alexander Fraser,Annemarie Friedrich
机构: TU Munich(慕尼黑工业大学); Munich Center for Machine Learning (MCML)(慕尼黑机器学习中心); Robert Bosch GmbH(罗伯特·博世公司); University of Augsburg(奥格斯堡大学); Munich Data Science Institute(慕尼黑数据科学研究所); LMU Munich(慕尼黑路德维希-马克西米利安大学)
类目: Computation and Language (cs.CL)
备注:
Abstract:A key factor in deciding whether to trust an automatic prediction is its confidence score, which should be calibrated to match the actual probability of the prediction being correct. Most confidence calibration metrics target binary or multi-class tasks, while multi-label calibration remains largely underexplored. Multi-label classification tasks, such as assigning medical codes to clinical notes or determining news topics, are usually dominated by a large number of negatives, i.e., labels that do not apply. We show that existing binning schemes to compute label-wise expected calibration error either underestimate the error, simply reflect label frequency, or suffer from many bins with very few instances. To achieve trustworthy label-wise calibration errors, we propose a new binning scheme that gives equal weight to positive and negative label assignments. Our empirical study demonstrates that in contrast to existing binning schemes, our new scheme results in meaningful estimates of calibration error in hierarchical and in extreme multi-label classification. We also show that calibrating confidence scores of large language models for multi-label predictions is an open challenge. Our detailed analysis lays the foundation for further research by providing a solid evaluation metric for measuring calibration in multi-label classification.
[NLP-15] Enriching Speech Emotion Representations with Conversational Context ICASSP2027
【速读】： 该论文旨在解决语音情感识别（Speech Emotion Recognition, SER）在实际对话场景中因忽略对话上下文而导致的情感演化捕捉不足的问题。现有方法多局限于单句级情感预测，未能充分建模对话过程中的情感流动与说话人互动关系。为此，本文提出ACERT（Averaged Contextual Emotion Representation through Time）模块，通过引入可变长度的对话上下文窗口，有效捕捉语音交互中情感随时间演进的动态特性。其核心创新在于利用时序上平均化的上下文情感表示，增强模型对情感连续性与对话连贯性的建模能力。实验结果表明，ACERT在IEMOCAP数据集上超越当前最优方法，在SAFE数据集上建立了首个面向上下文感知的基准，并在MELD数据集上实现了未加权与类别平衡指标下的优异表现。消融研究进一步验证，性能提升主要源于情感与对话连续性建模，而非说话人身份或声学条件等干扰因素。
链接: https://arxiv.org/abs/2609.26422
作者: Arthur Peuvot,Romaric Besançon,Gaël de Chalendar,Bianca Vieru,Ioana Vasilescu
机构: CEA List FactoryIA supercomputer (CEA 列表工厂IA超算中心); Ile-de-France Regional Council (法兰西岛大区议会); ANR-23-PEIA0008 SHARP project (ANR-23-PEIA0008 SHARP 项目); France 2030 program (法国2030计划)
类目: Computation and Language (cs.CL); Audio and Speech Processing (eess.AS)
备注: 5 pages, 1 figure, 2 tables. Submitted to ICASSP 2027
Abstract:Detecting emotions is necessary for building systems that can accurately and adaptively interact with humans. Speech Emotion Recognition (SER) has become an important research focus to develop intelligent spoken interfaces. However, most studies predict emotions at the utterance level, ignoring the conversational context, along with the emotional flow and speaker interactions it carries. In this paper, we introduce ACERT (Averaged Contextual Emotion Representation through Time), a module that integrates a flexible-length window of conversational context to better capture emotional evolution in spoken interactions. To evaluate the robustness of this method, we conducted experiments on datasets spanning diverse emotionally expressive styles and contexts. ACERT outperforms current state-of-the-art (SOTA) approaches on IEMOCAP, establishes the first context-aware benchmark on SAFE, and obtains strong results on MELD for unweighted, class-balanced metrics. Ablation studies show that ACERT’s gains come from emotional and conversational continuity, rather than from speaker identity or acoustic conditions.
[NLP-16] Combining Hierarchical Cognitive Process with Process Supervision for Interpretable Scene Safety Understanding
【速读】： 该论文旨在解决场景安全理解（scene safety understanding）在关键领域中因传统方法依赖直接映射而缺乏可解释性，从而限制其在高可靠性应用中的可信度的问题。其核心解决方案是通过建模人类认知过程，将场景安全的认知机制融入机器学习框架，构建一个分层的认知安全结构，并基于多步推理与过程标签构建高质量的场景安全理解数据集。在此基础上，提出一种模块化、灵活的过程监督框架，利用大语言模型（LLM）作为核心架构，结合低秩适配（LoRA）与专家混合（MoE）策略，实现多个专家模块在推理链中对特定子任务的分工协作与协同优化。该框架不仅提升了模型的性能与可解释性，还支持通过信息流与显著性分析等技术对中间推理步骤进行精细化解析，从而实现更透明、可靠的场景安全判断。
链接: https://arxiv.org/abs/2609.26399
作者: Zhiyun Jiang,Hanyong Wang,Binbin Liang,Yu Xie,Zhengjie Wang,Menglong Yang,Wei Li
机构: 未知
类目: Computation and Language (cs.CL)
备注:
Abstract:Scene safety understanding plays a life-or-death role in situational awareness in various critical domains. Traditional methods that rely on learning direct mappings between scenes and safety levels often lack interpretability, limiting their reliability in critical applications. An effective approach to overcoming this challenge lies in interpreting human cognitive processes and equipping machine models with analogous cognitive capabilities. This work explores an effective way of integrating scene safety cognitive process modeling and process supervision. Specifically, we first construct a hierarchical cognitive safety structure, which motivates the development of a novel, high-quality scene safety understanding dataset based on multi-step reasoning with process labels. This dataset serves both as a benchmark and a resource to improve the safety reasoning capabilities of Large Language Models (LLMs), while also enabling a granular analysis of intermediate reasoning steps through information flow and saliency-based techniques. Building upon this foundation, we introduce a modular and flexible process supervision framework that reflects the hierarchical nature of human cognition. This framework leverages LLMs as the core architecture and incorporates Low-Rank Adaptation(LoRA) and Mixture-of-Experts (MoE) strategies to enable specialization and collaboration among expert modules, each tasked with specific sub-processes of the overall reasoning chain. Systematic experimental evaluations and analyses confirm that our framework exhibits superior interpretability and performance characteristics compared to traditional approaches.
[NLP-17] On the Lexical Superstition of Large Language Models for Code Comprehension: Re-evaluation on Code of Low Lexical Quality
【速读】： 该论文旨在解决当前大语言模型（Large Language Models, LLMs）在代码理解任务中对标识符名称（identifier names）的过度依赖问题。尽管自然代码中的标识符名称具有统计信息价值，但其语义可能不可靠，而现有模型在重命名时往往过分依赖词汇线索（lexical cues），导致对程序结构的正确理解被误导。论文提出一种语义保持的标识符重命名框架 Face/Off，通过逐步去除或篡改标识符信息来评估多个模型在不同代码理解任务中的表现。结果显示，几乎所有测试模型均表现出显著的词汇过强调倾向：随着标识符信息被移除或故意误导，模型性能下降，且输出结果倾向于受误导性名称所暗示的错误语义。这一现象在提示工程（prompt-based）和微调（fine-tuning）等主流干预手段下依然存在，表明该问题具有深层结构性根源。进一步的类型推断控制实验揭示了关键边界：当目标答案可通过局部上下文独立恢复而无需依赖标识符名称时，命名干扰效应显著减弱。因此，解决方案的关键在于识别并缓解模型对表面词汇特征的非理性依赖，推动建立既能利用自然代码规律优势，又能基于精确形式化语义进行推理的评估与建模方法。
链接: https://arxiv.org/abs/2609.26388
作者: Xin Shen,San-Zhuo Xi,Yali Du,Ming Li(Nanjing University, Nanjing, China)
机构: Nanjing University (南京大学)
类目: oftware Engineering (cs.SE); Computation and Language (cs.CL); Machine Learning (cs.LG)
备注: 27 pages, 9 figures, 12 tables. Submitted to an ACM journal in September 2025. Preprint; manuscript under review. Corresponding author: Ming Li
Abstract:Recent advances in large language models (LLMs) have made them widely used for code-related tasks. Identifier names are statistically informative in naturally occurring code, but their information is not always reliable. We investigate whether current LLMs assign disproportionate weight to lexical cues when renaming preserves program structure. We introduce Face/Off, a semantics-preserving identifier-renaming framework, and evaluate progressive naming conditions across multiple models and code-comprehension tasks. Within this framework, lexical overemphasis is pervasive across the evaluated models and primary tasks: performance generally decreases as identifier information is removed or made misleading, and outputs are often directed toward the meanings suggested by misleading names. The pattern persists under representative prompt- and fine-tuning-based interventions, suggesting that lexical overemphasis is an entrenched problem. A type-inference control confirms a boundary: naming effects are smaller when the answer is locally recoverable without the target name. These results do not imply that identifiers are unhelpful; rather, they reveal a systematic vulnerability in how current LLMs balance lexical cues against program structure. Our findings motivate evaluations and modeling methods that preserve the benefits of natural code regularities while keeping conclusions grounded in accurate, formalized code semantics.
[NLP-18] Layout-Guided Masking for GROBID: Lightweight Structural Gains in Large-Scale Scientific PDF Ingestion
【速读】： 该论文旨在解决科学文献PDF转换为可机器读取全文时的结构化瓶颈问题，尤其针对现有视觉模型依赖GPU、计算成本高且易引入噪声的局限性。其核心解决方案是将轻量级CPU运行的检测器与现有的GROBID（一种基于字体流的模块化解析器）相结合：通过生成带类型标签的区域掩码（typed-area masks）来定位图表、表格及页眉页脚等副文本区域，这些区域的文本片段被定向至GROBID的专用模型进行处理或直接丢弃，从而实现高效精准的结构化提取。实验表明，在两个PMC数据集（生物信息学与材料科学）上，该方法在多数指标上优于纯GROBID，特别是在段落召回率和图注关联的图文恢复方面表现显著提升；在外部Table-BRGM基准测试中，表格检测F1值从0.16大幅提升至0.94，结构解析性能亦明显改善。同时，在主体文本处理方面，该方案在段落精度和章节识别上优于四种主流视觉模型，并保持接近最优GPU解析器的字符错误率。整个系统全程在CPU上运行，相比最便宜的GPU系统（Docling）节省2.7–3.2倍开销，较生成式解析器降低10–14倍成本，展现出卓越的效率与可扩展性。
链接: https://arxiv.org/abs/2609.26381
作者: Luca Foppiano,Sana Khamassi,Vipul Gupta
机构: ScienciaLAB(葡萄牙); Helmholtz-Zentrum Hereon(德国)
类目: Computation and Language (cs.CL); Digital Libraries (cs.DL)
备注:
Abstract:Transforming scholarly PDFs into machine-readable fulltext remains a bottleneck for large-scale information systems. Recent vision-based parsers improve accuracy, but need GPUs and may introduce noise into the extracted text. GROBID, a modular font-stream parser running on CPU, is the de-facto standard for structuring scientific articles and underpins several of the largest open scholarly corpora. We pair it with a lightweight CPU detector localising figure, table, and paratext (header, footer, page number) regions, encoded as typed-area masks whose tokens are routed to GROBID’s specialised models or discarded. On two PMC corpora, Bioinformatics (1,926 articles) and Materials Science (2,595), scored against JATS with a section-aware structural protocol, our extension improves over plain GROBID on most metrics (NS +0.025 / +0.013 ; +0.086 paragraph recall on Materials Science, d_z=1.08 ), and caption-linked figure recovery improves on both corpora. On the external Table-BRGM benchmark, table detection recovers F1 0.16 \to 0.94 and table structure follows (GriTS-Top 0.27 \to 0.78 , below the strongest GPU system). On body text, against four vision-based systems (Docling, MinerU, olmOCR, this http URL), it has the best paragraph precision on both corpora, the best section detection on Materials Science, and a character error rate within 0.004 of the best GPU parser. End-to-end on CPU, it costs 2.7 – 3.2\times less than the cheapest GPU system (Docling) and 10 – 14\times less than generative parsers.
[NLP-19] HySparse2: Hybrid Sparse Attention with Two-Level KV Sharing
【速读】： 该论文旨在解决长时序、多轮交互智能体在生成短动作、处理长观测序列时面临的上下文管理挑战，具体包括高效预填充（prefill）、紧凑的键值缓存（KV-cache）存储以及准确的长上下文检索问题。其核心解决方案是提出一种两级共享键值缓存（two-level KV sharing）的混合稀疏注意力架构HySparse2：在外部层级，采用类YOCO的自解码器与交叉解码器结构，仅对全注意力层进行桥接，其中自解码器使用混合滑动窗口注意力（hybrid sliding-window attention, SWA），交叉解码器则采用混合稀疏注意力，且交叉解码器的KV缓存由自解码器中全注意力层的隐藏状态生成；在内部层级，保留HySparse的核心KV重用设计，并引入两项改进：将块级稀疏性替换为粒度更细的令牌级稀疏性以提升长上下文检索精度，同时取消稀疏层中的独立滑动窗口分支，强制最近令牌进入稀疏选择范围。这一两级共享机制使得所有交叉解码器的KV缓存均可由自解码器隐藏状态构建，从而实现预填充阶段仅需执行自解码器即可完成，跳过全部交叉解码器层。实验表明，在80B-A3B MoE模型上，HySparse2在长上下文检索与多轮代理任务中优于HySparse与Hybrid SWA，同时显著降低预填充计算开销与KV缓存存储需求。
链接: https://arxiv.org/abs/2609.26368
作者: Jianyu Wei,Yizhao Gao,Qihao Zhang,Shimao Chen,Zhengju Tang,Yu Cheng,Shengjie Zhou,Zihan Jiang,Yifan Song,Hailin Zhang,Liang Zhao,Bo Yang,Gang Wang,Shijie Cao,Fuli Luo
机构: 未知
类目: Computation and Language (cs.CL)
备注:
Abstract:Long-horizon and multi-turn agents typically generate short actions and process long observations from tools and environments. This growing context demands efficient prefill, compact KV-cache storage, and accurate long-context retrieval. To meet these demands, we introduce HySparse2, a hybrid sparse attention architecture with two-level KV sharing. At the outer level, KV Bridging adopts a YOCO-style self-decoder and cross-decoder structure, but bridges only full-attention layers. The self-decoder uses hybrid sliding-window attention (SWA), while the cross-decoder uses hybrid sparse attention. The KV caches for full-attention layers in the cross-decoder are generated from the hidden states of full-attention layers in the self-decoder. At the inner level, HySparse2 retains HySparse’s core KV Reuse design with two refinements. First, it replaces block-level sparsity with token-level sparsity for finer long-context retrieval. Second, it removes the separate SWA branch from sparse layers and instead forces a sliding window of recent tokens into the sparse selection. This two-level KV sharing allows all cross-decoder KV caches to be constructed from self-decoder hidden states. Prefill can therefore exit after the self-decoder, skipping all cross-decoder layers. On an 80B-A3B MoE model, HySparse2 outperforms HySparse and Hybrid SWA on long-context retrieval and multi-turn agentic tasks, while substantially reducing prefill computation and KV-cache storage.
[NLP-20] ransBERT: A Framework for Synthetic Translation in Domain-Specific Language Modeling
【速读】： 该论文旨在解决专业领域中非英语语料稀缺问题，这一瓶颈严重制约了自然语言处理（Natural Language Processing, NLP）工具在低资源语言-领域组合中的有效开发。其核心解决方案在于提出TransBERT框架，通过仅使用合成翻译文本进行预训练语言模型的构建，并配套推出可扩展的TransCorpus翻译工具包。研究聚焦于法语生命科学领域，实证表明仅依赖合成翻译数据即可在多个下游任务上实现前沿性能。关键突破在于验证了在高资源翻译方向（如英→法）中，利用高质量合成翻译文本可有效构建低资源语言-领域组合下的高质NLP资源，从而为跨语言NLP发展提供了可行路径。
链接: https://arxiv.org/abs/2609.26347
作者: Julien Knafou,Luc Mottin,Anaïs Mottaz,Alexandre Flament,Patrick Ruch
机构: HES-SO(瑞士高等教育与研究学院); SIB(瑞士生物信息学研究所)
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
备注: 17 pages
Abstract:The scarcity of non-English language data in specialized domains significantly limits the development of effective Natural Language Processing (NLP) tools. We present TransBERT, a novel framework for pre-training language models using exclusively synthetically translated text, and introduce TransCorpus, a scalable translation toolkit. Focusing on the life sciences domain in French, our approach demonstrates that state-of-the-art performance on various downstream tasks can be achieved solely by leveraging synthetically translated data. We release the TransCorpus toolkit, the TransCorpus-bio-fr corpus (36.4GB of French life sciences text), TransBERT-bio-fr, its associated pre-trained language model and reproducible code for both pre-training and fine-tuning. Our results highlight the viability of synthetic translation in a high-resource translation direction for building high-quality NLP resources in low-resource language/domain pairs.
[NLP-21] Blaming Across the Aisle: Political Contrasting and Blame Attribution in the Danish Parliament ACL COLING2027
【速读】： 该论文旨在解决政治话语日益敌对化这一广泛感知现象背后的实证证据不足问题，聚焦于丹麦议会1997至2026年间责任归因（blame attribution）的演变趋势。其核心解决方案在于构建一种面向低至中等资源语言的高效标注流程，并开发名为BlameBERT的专用分类器（F1值达0.80），结合多层次统计建模方法，系统分析政治话语中的责任归因模式。研究发现，责任归因呈现出“香蕉形”轨迹——先持续下降至约2016年，随后在2019至2026年间出现显著且持续上升。政府地位始终显著影响责任归因，形成所谓的“政治对比效应”（political contrasting），即反对党比执政党更倾向于归责；该效应受意识形态调节：右翼政党执政时的责任抑制作用较弱，且意识形态极端性在右翼群体中更强烈地放大了责任归因。近年来，政治光谱右翼内部政党在意识形态极端性与政治立场之间的交互作用显著增强，表明政治话语中的归责修辞呈现以右翼为中心的意识形态固化趋势。综合来看，研究揭示了所谓政治语言日趋严厉并非单纯的修辞漂移，而是具有意识形态非对称性的深层话语硬化现象。敏感性分析进一步验证了结论对分类阈值变化的稳健性。
链接: https://arxiv.org/abs/2609.26346
作者: Markus Lundsfryd Jensen,Rune Egeskov Trust,Kenneth Christian Enevoldsen,Sara Kolding
机构: Aarhus University (奥胡斯大学); Center for Humanities Computing (人文计算中心)
类目: Computation and Language (cs.CL)
备注: 8 Pages + appendix (25 total) Main paper 4 figures 2 tables: Appendix 9 figures 10 tables. Model found here: this https URL , dataset here: this https URL . Markus Lundsfryd Jensen and Rune Egeskov Trust have contributed equally. Paper will be submitted through ACL rolling review (ARR), we are aiming for COLING 2027
Abstract:Political discourse is widely perceived to be growing more hostile, yet robust evidence remains scarce. This study examines blame attribution in the Danish Parliament from 1997 to 2026, combining a purpose-built classifier, BlameBERT (F1: 0.80), with multilevel statistical modeling. The classifier is constructed using an annotation-efficient pipeline for blame attribution in low-to-mid resource languages. The results reveal a banana-shaped trajectory, with blame declining until around 2016 before entering a significant and sustained increase in recent years (2019-2026). Government status consistently influenced blame attribution - an effect we term political contrasting - with opposition parties blaming substantially more than governing parties. This effect was moderated by ideology: The blame-dampening effect of governing was less pronounced among right-wing parties, and ideological extremity amplified blame more strongly on the right. In recent years, the interaction between political wing and ideological extremity intensified, suggesting an ideological hardening of the blame rhetoric concentrated on the right of the political spectrum. Taken together, these patterns suggest that the perceived rise in harsh political language reflects not merely a general rhetorical drift, but an ideologically asymmetric hardening of political discourse. A sensitivity analysis showed that the conclusions were robust to varying classification thresholds.
[NLP-22] Designing and Analysing Argument Mining Pipelines: Towards a Comprehensive Assessment
【速读】： 该论文旨在解决当前论点挖掘（Argument Mining, AM）研究中因不同方法对任务概念化不一致而导致的可比性差、评估不透明的问题。现有端到端AM流水线在任务划分与实现方式上缺乏统一标准，使得跨方法比较困难。为此，本文提出一个三重视角分析框架——语言学视角、计算视角与领域视角，系统地剖析多个前沿端到端AM方法的流水线设计，以揭示其如何建模论点结构、进行计算处理以及融合领域知识。该研究的关键在于构建了一个通用的设计范式，从语言学与计算两个维度阐明核心论点挖掘任务的建模与实现机制，从而为未来AM方法提供以方法论为中心的标准化描述基础，推动更深入的理解与系统性比较。
链接: https://arxiv.org/abs/2609.26338
作者: Siddharth Bhargava,Sara Tonelli,Patricia Martín-Rodilla
机构: Fondazione Bruno Kessler(布鲁诺·凯斯勒基金会), Trento, Italy(意大利特伦托); Universidade da Coruña(拉科鲁尼亚大学), A Coruña, Spain(西班牙拉科鲁尼亚); IEGPS-CSIC(西班牙国家研究委员会艾格普斯研究所), Santiago de Compostela, Spain(西班牙圣地亚哥德孔波斯特拉)
类目: Computation and Language (cs.CL)
备注: 12 pages, 3 figures, European Conference on Argumentation 2025 (ECA 2025)
Abstract:Argument Mining (AM) transforms natural language into its underlying argument structures. This transformation is typically realized through a sequence of AM tasks that form an end-to-end AM pipeline. However, AM approaches often differ in how they conceptualize these tasks, making direct comparisons between them difficult and opaque. This calls for a more nuanced, task-level analysis of AM approaches to enable clearer comparison and assessment. This work presents a preliminary meta-study that systematically reviews several state-of-the-art end-to-end AM works and analyzes their pipelines through a triple-perspective framework—a linguistic, computational and domain perspective—to understand how the pipelines model arguments as structures, computes them, and integrates domain knowledge. We further propose a general design to the linguistic and computational perspectives, illustrating how key AM tasks are designed for modeling and computation of argument structures. Our proposed framework lays the groundwork for methodology-centered descriptions across AM approaches, facilitating deeper understanding and more systematic comparisons in future research. Comments: 12 pages, 3 figures, European Conference on Argumentation 2025 (ECA 2025) Subjects: Computation and Language (cs.CL) Cite as: arXiv:2609.26338 [cs.CL] (or arXiv:2609.26338v1 [cs.CL] for this version) https://doi.org/10.48550/arXiv.2609.26338 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
[NLP-23] CHiME-9 ECHI: A Machine Learning Challenge for Enhancing Conversations to Address Hearing Impairment
【速读】： 该论文旨在解决在嘈杂的食堂环境背景下，四人对话中因听力障碍导致语音可懂度与质量下降的问题。具体而言，挑战任务聚焦于从多通道录音中分离出目标说话人的语音，以应对多个干扰性语音源和背景音效带来的挑战。其核心解决方案在于利用佩戴在Meta Aria眼镜及助听器上的麦克风采集的音频数据，结合干净语音样本，通过先进的语音增强算法实现对目标语音的有效提取。关键创新点在于融合多源音频信息并提升语音的可懂度与感知质量，尽管客观评价指标未能完全反映听觉表现，但排名靠前的系统在主观听感测试中显著优于基线模型，在语音清晰度与整体质量方面均取得实质性提升。
链接: https://arxiv.org/abs/2609.26306
作者: Robert Sutherland,Thomas Kuebert,Marko Lugger,Stefan Petrausch,Eline Borch Petersen,Juan Azcarreta Ortiz,Buye Xu,Stefan Goetze,Jon Barker
机构: 未知
类目: Computation and Language (cs.CL)
备注: Accepted to the International Workshop on Acoustic Signal Enhancement (IWAENC), Cremona, Italy, September 2026
Abstract:This work presents the task and results of the CHiME-9 challenge for Enhancing Conversations to address Hearing Impairment. The challenge considers the scenario of four-party conversations in a noisy, cafeteria-style environment with interfering speech sources and sound effects. Participants are provided with audio recordings made with Meta Aria glasses and hearing aid microphones, and clean speech samples of the conversation participants. The task is to extract the speech of the conversation partners from the noisy multi-channel recordings with the goal of improving the intelligibility and quality of the speech, evaluated using objective metrics and subjective listening tests. This paper reviews submissions from seven teams and ranks them on a combination of subjective intelligibility and quality. Results show that while the objective metrics do not reflect listener performance, the top systems were able to make substantial improvements over the challenge baseline in both intelligibility and quality ratings.
[NLP-24] PACE-dLLM : Elastic Block Decoding via Confidence Cliff Estimation for Diffusion Language Models
【速读】： 该论文旨在解决生成式 AI（Generative AI）中扩散语言模型（Diffusion Language Models, dLLMs）在推理加速过程中存在的关键瓶颈问题：现有块级解码（block-wise decoding）策略中，解码块长度 $ B $ 同时耦合了前瞻视野（look-ahead horizon）与待确认词元数量两个独立决策，导致难以实现高效且精准的并行生成。传统加速方法依赖间接启发式规则（如波动性追踪、分隔符检测或学习评分函数），缺乏对模型内部置信度信息的直接利用。本文提出 PACE-dLLM，其核心创新在于发现模型每步预测的局部置信度（in-window confidence）呈现一种上下文相关的“悬崖效应”（cliff），其饱和点直接指示最优前瞻视野。通过在每一步以闭式形式拟合该参数化悬崖，PACE-dLLM 依据饱和点动态设定下一阶段的前瞻范围，并采用独立的置信阈值进行词元提交。在饱和收益（saturated-yield）抽象下，证明该悬崖锚定的前瞻视野是达到最大单位前向传递有用产出的最小有效视野：低于此视野将导致更差的渐近非函数评估次数（NFE）率，而高于则不增加有效产出。在四个推理与代码生成基准测试中，PACE-dLLM 在开源 dLLM 基础架构上均取得最佳平均准确率，相较未加速的半自回归基线，在 LLaDA 上实现平均 5.23× 的墙时速度提升，在 Dream 上达 3.06×（数学任务最高达 8.52×），显著推进了生成质量与吞吐量之间的帕累托前沿。
链接: https://arxiv.org/abs/2609.26249
作者: Xiaocheng Lu,Shuhan Guo,Ziyue Ma,Jie Zhang,Jian Liu,Jingcai Guo,Haoxuan Che,Song Guo
机构: The Hong Kong University of Science and Technology(香港科技大学); Northwestern Polytechnical University(西北工业大学); The Hong Kong Polytechnic University(香港理工大学)
类目: Computation and Language (cs.CL)
备注: 11 pages, 5 figures, 2 tables
Abstract:Diffusion language models (dLLMs), such as LLaDA and Dream, have become competitive with autoregressive (AR) LLMs in generation quality while supporting native parallel decoding. A standard acceleration strategy is block-wise decoding, where each forward pass predicts a block of length B and commits high-confidence tokens. However, B couples two distinct decisions: the look-ahead horizon and the number of tokens to commit. Existing accelerators address this limitation through indirect heuristics, such as volatility tracking, delimiter detection, and learned scoring. In contrast, we show that the required information is already encoded in the model’s own per-step confidence: in-window confidence typically follows a context-dependent cliff, whose saturation point directly identifies the appropriate look-ahead horizon. We propose PACE-dLLM, which fits this parametric cliff in closed form at each step, sets the next horizon by its saturation point, and uses an independent confidence threshold for token commitment. Under a saturated-yield abstraction, we show that the cliff-anchored horizon is the smallest horizon attaining maximal useful per-pass yield: fixed horizons that undershoot it incur a worse asymptotic NFE rate, while overshooting adds no useful yield. On four reasoning and code benchmarks, PACE-dLLM achieves the best average accuracy on both open-source dLLM backbones, with average wall-clock speedups of 5.23x on LLaDA and 3.06x on Dream (up to 8.52x on math) over the unaccelerated semi-AR baseline, advancing the quality-throughput Pareto frontier.
[NLP-25] Damage Predicts Recovery: When Calibration Data Matters in Compressing Financial LLM s
【速读】： 该论文旨在解决生成式AI在特定领域（如金融）中进行后训练量化与剪枝压缩时，是否需要使用领域匹配的校准数据这一未决问题。其核心观点是：是否需要领域专用校准数据，取决于压缩操作对目标任务造成的性能损伤程度，而非数据领域的不匹配本身。若压缩操作基本保留目标能力，则校准数据的选择影响较小；若压缩导致显著性能下降，则使用任务格式化的领域专用校准数据（如FinMix，一种金融任务样本混合数据）可有效恢复部分损失。研究通过在两个模型家族、六种压缩配置、三种词元匹配的校准语料及十项金融分类与数值问答任务上的实验验证了该假设。结果表明，量化对任务性能影响较小，校准数据选择无关紧要；而剪枝使数值问答准确率下降超过40个百分点，在此受损情况下，通用校准数据无法缓解损失，而基于金融任务的FinMix则显著恢复性能。该结论在不同模型架构和规模下均成立，支持了一条实用准则：应优先评估压缩带来的任务特异性损伤，仅当损伤严重时才构建领域专用校准数据。
链接: https://arxiv.org/abs/2609.26241
作者: Junyi Ye,Mengjia Yu,Debapriya Hazra,Guiling Wang
机构: Montclair State University (蒙特克莱尔州立大学); New Jersey Institute of Technology (新泽西理工学院)
类目: Computation and Language (cs.CL); Machine Learning (cs.LG)
备注:
Abstract:Post-training quantization and pruning rely on a small calibration corpus. Whether specialized domains such as finance require domain-matched calibration data remains unsettled. We argue that the answer depends on the task-level damage caused by compression rather than on domain mismatch. If compression preserves the target capability, changing the calibration corpus has little effect. If compression causes large losses, task-formatted calibration can recover part of the loss. We test this hypothesis across two model families, six compression configurations, three token-matched calibration corpora, and ten financial classification and numerical question-answering tasks. The results support this hypothesis. Quantization largely preserves task performance, and calibration choice has little effect in this case. Pruning reduces numerical QA accuracy by over 40 points. In these damaged settings, another generic corpus does not help, while FinMix, a mixture of financial task examples, recovers a large part of the loss. The link between damage and recovery holds across model families and scales. These findings support a practical rule. Measure task-specific compression damage first, and construct specialized calibration data only when the damage is large.
[NLP-26] Same Chart Different Story: Bias in Vision-Language Chart Interpretation
【速读】： 该论文旨在解决生成式视觉语言模型（VLMs）在图表解释中可能产生的社会维度偏见问题，尤其关注当仅改变所提及的社会群体时，模型会生成不同甚至强化刻板印象的叙事，从而误导关键决策。其核心挑战在于现有方法缺乏系统性评估工具来检测模型在种族、收入、年龄、宗教、移民身份和性别等多维社会属性下的解释一致性与公平性。为此，论文提出了首个专门针对图表解释偏见的基准测试——ChartBias，包含820张真实世界图表，覆盖六大社会属性，共生成4,319个有效图表-属性实例及8,638对固定图表但仅更换群体术语的成对生成结果。实验发现，主流12种商业与开源VLM普遍存在三种典型失败模式：叙事漂移（相同图表生成不同叙述）、群体幻觉（无证据地将图表归因于特定群体）以及偏好极性（倾向将积极趋势关联至某一特定群体）。为缓解上述问题，研究提出一种多智能体协同的缓解框架，通过分离基于图表的证据提取与群体条件化生成过程，并引入反事实判别器验证群体驱动差异是否具有图表依据，显著降低叙事漂移，同时保持对图表内容的忠实推理。研究表明，评估图表理解能力不仅需关注准确性，更应综合考量跨社会群体的公平性与一致性。相关数据集已公开发布。
链接: https://arxiv.org/abs/2609.26210
作者: Mizanur Rahman,Huan Wu,Arash Asgari,Enamul Hoque Prince,Laleh Seyyed-Kalantari
机构: York University ( York 大学); Vector Institute (向量研究所); Connected Minds (连接心智)
类目: Computation and Language (cs.CL)
备注:
Abstract:Vision-language models (VLMs) are increasingly used to interpret charts and generate natural-language explanations for socially consequential data. However, they may produce different narratives for the same chart when only the referenced social group changes, reinforcing stereotypes and misleading decisions. Despite these risks, no benchmark exists for systematically evaluating bias in chart interpretation across social dimensions. We introduce ChartBias, the first benchmark for auditing bias in VLM-based chart interpretation. ChartBias contains 820 manually curated real-world charts spanning six attributes: race, income, age, religion, immigration status, and gender, yielding 4,319 valid chart, attribute instances and 8,638 paired generations where the chart is fixed and only the group term is swapped. Across 12 proprietary and open-source VLMs, totaling 155,484 model responses, we find three widespread failure modes: narrative shift (same chart, different narratives), group hallucination (assigning a chart to a group without evidence), and preference polarity (favourable trends often linked to one group). We further propose a multi-agent mitigation framework that serves as a strong baseline by separating chart-grounded evidence extraction from group-conditioned generation and using a counterfactual judge to verify that group-driven differences are supported by the chart. The framework substantially reduces narrative shift while preserving chart-grounded reasoning. Our findings show that evaluating chart understanding requires measuring not only accuracy, but also fairness and consistency across social groups. We release ChartBias at this https URL.
[NLP-27] Beyond Static Charts: Can Language and Vision Language Models Generate Interactive Data Visualization Interfaces?
【速读】： 该论文旨在解决生成式人工智能（Generative AI）在语言驱动的交互式数据可视化界面生成方面能力不足的问题。尽管现有大语言模型（LLM）和视觉语言模型（VLM）已能在自然语言指令下生成静态图表，但其在生成动态、可交互的数据可视化界面方面仍缺乏系统性评估与有效方法，主要受限于缺少高质量基准测试集。为此，论文提出了VIS-GEN基准，包含3,042个涵盖数据过滤、时间序列分析、可视化编辑等多样化分析意图的真实场景样本，每条样本均配有数据集元信息与自然语言查询，以模拟实际数据分析中的目标导向探索过程。实验表明，当前14个主流开源与闭源模型在涉及隐含意图、多交互路径及复杂编辑操作的任务中普遍存在性能瓶颈，凸显了交互式界面生成作为超越静态图表合成的关键挑战。为应对该问题，论文提出一种结构化的多阶段界面生成框架，将任务分解为可视化设计表示、多候选界面生成、约束感知批判与自精炼四个阶段，显著提升了模型表现，使最优模型通过率提升15.9个百分点，验证了该方法在构建更可靠语言驱动交互式可视化系统方面的可行性。
链接: https://arxiv.org/abs/2609.26208
作者: Mizanur Rahman,Aaryaman Kartha,Enamul Hoque Prince
机构: York University ( York 大学)
类目: Computation and Language (cs.CL)
备注:
Abstract:Data visualization is central to analytical reasoning, but real-world analysis increasingly requires language-driven interactive interfaces rather than static charts. Although recent large language and vision language models (LLMs/VLMs) have shown promise in generating static charts from natural language, their ability to generate interactive data visualization interfaces remains largely unexplored due to the lack of benchmarks. We introduce VIS-GEN, a benchmark for evaluating how well LLMs/VLMs can generate interactive visualization interfaces from natural language queries. VIS-GEN comprises 3,042 samples covering diverse analytical intents, including data filtering, temporal analysis, and visualization editing, each paired with dataset metadata and natural language queries that are designed to reflect realistic, goal driven data exploration scenarios. We benchmark 14 state-of-the-art open-source and closed-source LLMs/VLMs, revealing large performance gaps and frequent failures on queries involving implicit intent, multiple interaction alternatives, and complex editing operations, highlighting interactive interface generation as a key open challenge beyond static chart synthesis. To address this, we propose a structured multi stage interface generation framework that decomposes the task into visualization design representation, generation of multiple interface candidates, constraint-aware critique, and self-refinement. This approach improves the best models pass rate by 15.9 percentage points, demonstrating a practical path toward more reliable language-driven interactive visualization systems. We release VIS-GEN at this https URL.
[NLP-28] WatchPoint: Executable User Feedback for Real-World Agent ic Web Development
【速读】： 该论文旨在解决现有编码代理（coding agents）在代码测试失败后缺乏有效反馈机制的问题，尤其针对其无法像真实开发者一样与运行中的应用进行交互、难以精准定位和修复复杂多文件Web项目中顺序依赖任务缺陷的局限性。其核心解决方案是提出WatchPoint——一个模拟用户行为的系统，通过生成并执行诊断脚本对正在运行的应用程序进行主动探测，从而获取结构化观测数据，指导编码模型进行有针对性的重试。与以往仅依赖截图、大语言模型评分或自然语言修正等静态反馈方式不同，WatchPoint实现了动态、交互式的自动化调试过程。研究基于Web-Bench基准测试集（包含50个多文件Web项目及1,000个顺序依赖任务），采用可验证的确定性端到端测试评估性能，结果显示WatchPoint成功恢复了57.6%的任务；并通过受控用户研究证实，人类测试者在相同任务上也达到54.5%的恢复率，表明该系统的仿真行为具备高度现实性，能够替代人工交互式测试。此外，研究还揭示了影响模拟用户反馈有效性的能力差距模式，明确了何时应提供反馈以及何时应避免误导性干预。
链接: https://arxiv.org/abs/2609.26204
作者: Guanqun Yang,Wei Yang,Xueqing Liu
机构: Stevens Institute of Technology ( Stevens Institute of Technology); University of Texas at Dallas (University of Texas at Dallas)
类目: oftware Engineering (cs.SE); Artificial Intelligence (cs.AI); Computation and Language (cs.CL)
备注:
Abstract:When a professional web developer’s code fails a test, they do not simply re-read the stack trace. They open the application in a browser, click buttons, inspect computed styles, and run diagnostic commands to understand what went wrong. Existing feedback mechanisms for coding agents rely on screenshots, LLM-as-a-judge scoring, or natural-language corrections, but few interact with the live application the way a developer would. We introduce WatchPoint, a simulated-user system that mimics real developer behavior by generating and executing diagnostic scripts against the running application, producing structured observations that guide the coding model’s retry. Unlike prior approaches that target single-file edits or evaluate using non-executable metrics, we operate on Web-Bench, a benchmark of 50 multi-file web projects comprising 1,000 sequentially dependent tasks, verified by deterministic end-to-end tests. WatchPoint recovers 57.6% of the tasks it diagnoses, and a controlled user study confirms the simulation’s realism: human testers achieve a comparable recovery rate (54.5%), providing evidence that automated diagnostic scripts can substitute for interactive human testing on sequential web development tasks. We further identify a pattern of capability gaps that governs when simulated-user feedback is helpful and when it should be withheld.
[NLP-29] Dynamic Deep Prompt Optimization for Defending Against Jailbreak Attacks on LLM s
【速读】： 该论文旨在解决大语言模型（Large Language Models, LLMs）在实际应用中面临的越狱攻击（jailbreak attacks）问题，即攻击者通过特定输入诱导模型生成有害或非预期内容，而现有防御方法存在成本高、易遗忘或适应性差等局限。针对传统静态提示优化方法（如固定前缀或后缀）无法应对多样化和动态演进的攻击这一关键缺陷，本文提出一种名为动态深度提示优化（Dynamic Deep Prompt Optimization, DDPO）的新方案。其核心创新在于利用目标模型自身的中间层作为特征提取器，通过一个轻量级多层感知机（multilayer perceptron）动态生成具有针对性的防御嵌入（defensive embeddings），并将其注入到后续的中间层中，实现输入依赖的自适应防御。该方法无需修改模型权重，具备高度灵活性与极低计算开销，在多种模型和攻击场景下均显著优于传统静态提示优化方法，尤其在弱对齐模型及语义模糊的良性提示识别任务中表现出色，有效区分真实有害请求与伪装成良性的恶意输入。
链接: https://arxiv.org/abs/2609.26185
作者: Doniyorkhon Obidov,Honggang Yu,Xiaolong Guo,Kaichen Yang
机构: 1. National University of Singapore (新加坡国立大学); 2. Tsinghua University (清华大学); 3. Chinese Academy of Sciences (中国科学院)
类目: Cryptography and Security (cs.CR); Artificial Intelligence (cs.AI); Computation and Language (cs.CL)
备注:
Abstract:Large Language Models (LLMs) demonstrate impressive capabilities across many applications but remain vulnerable to jailbreak attacks, which elicit harmful or unintended content. While model fine-tuning is an option for safety alignment, it is costly and prone to catastrophic forgetting. Prompt optimization has emerged as a promising alternative, yet existing prompt-based defenses typically rely on static modifications (e.g., fixed prefixes or suffixes) that cannot adapt to diverse and evolving attacks. We propose Dynamic Deep Prompt Optimization (DDPO), the first jailbreak defense based on deep prompt optimization. DDPO uses the target LLM’s own intermediate layers as feature extractors to dynamically generate defensive embeddings via a lightweight multilayer perceptron. These tailored embeddings are then injected into a subsequent intermediate layer, enabling an input-dependent defense without modifying the LLM’s weights. This design ensures high adaptability with minimal computational overhead. Experiments on a diverse set of models and attacks demonstrate that DDPO significantly outperforms static prompt optimization methods, particularly on weakly aligned models and when handling semantically ambiguous benign prompts, successfully distinguishing them from genuinely harmful requests. Subjects: Cryptography and Security (cs.CR); Artificial Intelligence (cs.AI); Computation and Language (cs.CL) Cite as: arXiv:2609.26185 [cs.CR] (or arXiv:2609.26185v1 [cs.CR] for this version) https://doi.org/10.48550/arXiv.2609.26185 Focus to learn more arXiv-issued DOI via DataCite (pending registration) Related DOI: https://doi.org/10.1609/aaai.v40i42.40887 Focus to learn more DOI(s) linking to related resources
[NLP-30] Modality-Gated Deep Adapters: Adding a Modality to a Frozen Embedding Model with Exact Preservation
【速读】： 该论文旨在解决多模态嵌入模型在规模化部署时，因新增模态而引入的输出不一致性问题。现有参数高效微调方法（如LoRA）在扩展新模态时会隐式修改基模型的输出，导致检索索引、基准测试结果和行为审计等依赖于固定嵌入输出的应用失效。其核心解决方案是提出模态门控深度适配器（modality-gated deep adapters）：在冻结的多模态嵌入大语言模型（Multimodal Embedding LLM）的每个解码器层上附加瓶颈适配器，并按模态分组，仅在对应模态编码时激活。该设计确保未被激活的模态输入完全沿用基模型原有计算图，实现比特级不变性；多个共加载适配器通过精确零隔离矩阵组合，保证各模态间无干扰。上述性质以数学命题形式严格定义，适用于任意训练后的状态，无需任务标签或推理时路由元数据，且经由释放检查点的精确相等性验证。实验表明，在一个20亿参数的冻结基模型上，音频适配器（通过连接符注入）使音频到文本的R@10提升3.4至5.4个百分点，且在所有种子下均显著优于对照组，复现效果在十倍数据量下依然成立；热成像适配器则利用基模型已冻结的视觉路径，将热成像输入的注册通过率提升近七倍，使热成像到文本的R@10从0.224提升至0.785。通过编码器替换实验进一步确认，模型容量缺失位于解码层内部，恰好由门控适配器所处位置填补。研究团队已开源音频模型、热成像适配器及完整的训练、评估与不变性验证工具链。
链接: https://arxiv.org/abs/2609.26182
作者: Abdul Basit Tonmoy,Kazi Fardinul Hoque,Md. Shahrier Islam Arham,Arman Luthra
机构: Eximius Labs(Eximius实验室); Wabash College(瓦巴什学院); Skop Intelligence Co.(Skop智能公司)
类目: Computation and Language (cs.CL); Computer Vision and Pattern Recognition (cs.CV); Sound (cs.SD)
备注: 14 pages, 3 figures, 6 tables. Models: this https URL and this https URL . Code: this https URL
Abstract:Multimodal embedding models are deployed at scale: retrieval indices, benchmark results, and behavioral audits all depend on the base model’s exact outputs. Extending such a model to a new modality with existing parameter-efficient methods silently changes those outputs; LoRA-style adaptation rewrites the text path whether or not the weights are merged, invalidating every stored embedding. We propose modality-gated deep adapters: bottleneck adapters attached to every decoder layer of a frozen multimodal embedding LLM, grouped into per-modality packs that execute only while their own modality is being encoded. The result is a modality added with zero change to existing outputs: inputs no pack claims traverse the base model’s own computation graph, bit-for-bit unchanged, and co-loaded packs compose with an exact-zero isolation matrix. Both properties are stated as propositions, hold after arbitrary training rather than only at initialization, require no task labels or routing metadata at inference, and are verified by exact-equality tests on the released checkpoints. On one frozen 2B base, the audio pack (injected as connector tokens) improves audio-to-text R@10 by +3.4 to +5.4 points over an identically trained control, positive at every seed and reproduced at eleven times the data; the thermal pack, reusing the base’s own frozen vision path, clears its pre-registered acceptance gate roughly sevenfold at every seed and lifts thermal-to-text R@10 from 0.224 to 0.785. An encoder swap locates the missing capacity: an external audio encoder that outranks Whisper-family encoders in CLAP-style comparisons loses by 16 R@10 points inside the frozen LLM, so the capacity belongs in the layers, exactly where the gated adapters place it. We release the audio model, the thermal pack, and the training, evaluation and invariance suites: models at this http URL, code on GitHub.
[NLP-31] Magnitude Profile Pruning: Calibration-Free Structured Attention Head Removal for Transformer Compression ICML
【速读】： 该论文旨在解决结构化剪枝（structured pruning）中注意力头重要性评估依赖校准数据、梯度计算或海森矩阵估计所带来的额外开销与数据依赖性问题。现有方法在实际应用中受限于对特定输入数据的敏感性及高昂的计算成本，难以实现高效、通用的模型压缩。本文提出一种无需训练的头重要性评分机制——权重行范数统计异常检测（Magnitude Profile, MP），通过分析注意力头投影权重的行范数分布，识别出处于总体分布内的冗余头（即“正常”范数头）进行剪枝，而保留具有显著异常范数、承载更强表征能力的头。进一步地，针对分组查询注意力（Grouped Query Attention, GQA）结构，提出MP-G变体，将共享键值组的得分合理分配至对应查询头，以保持剪枝精度。实验表明，在五种模型上，MP-G在12.5%至50%头稀疏度下均取得最优性能，如OPT-6.7B在50%稀疏度时困惑度仅为152.0，优于依赖校准的Wanda-Head、SparseGPT-Head和Gradient-Head等基线方法；在RoBERTa-large上亦表现领先。该方法不需前向传播、校准样本或梯度计算，实现零成本剪枝，同时在50%稀疏度下可带来最高达16%的参数缩减与50%的注意力计算量降低。研究结果证明，仅基于权重的统计评分即可达到甚至超越依赖数据的方法，为Transformer模型压缩提供了一种高效、实用且可扩展的无监督剪枝方案。
链接: https://arxiv.org/abs/2609.26177
作者: Kasun Dewage,Marianna Pensky,Heranga K. Rathnasekara,Suranadi De Silva
机构: University of California, Santa Barbara (加州大学圣塔芭芭拉分校); University of Colombo (科伦坡大学)
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
备注: Accepted as a regular paper at IEEE ICMLA 2026; to appear in the conference proceedings
Abstract:Structured pruning of attention heads provides a hardware-friendly way to compress Transformer language models. However, existing methods for measuring head-level importance require calibration data, gradient computation, or Hessian estimation. These requirements add extra overhead and make the methods depend on the data. Our work presents Magnitude Profile (MP) scoring, a training-free criterion for head importance that identifies dispensable heads through statistical outlier detection on weight row norms. Heads whose projection weights fall within the population bulk are pruned, while heads exhibiting outlier norms, which carry disproportionate representational capacity, are preserved. Our work further gives MP-G, a variant that handles Grouped Query Attention (GQA) by distributing shared key-value group scores across associated query heads. Across five models evaluated on WikiText-2 perplexity at 12.5%-50% head sparsity, MP-G achieves the best perplexity on OPT-6.7B at all sparsity levels (18.46 at 12.5%, 27.87 at 25%, 152.0 at 50%). MP-G also gives the best results on RoBERTa-large at 12.5% and 25% sparsity, with perplexity values of 7.27 and 10.28, outperforming calibration-dependent baselines including Wanda-Head, SparseGPT-Head, and Gradient-Head. It requires zero forward passes, calibration samples, or gradient computation. At 50% sparsity, head pruning yields up to 16% parameter reduction with 50% attention FLOP savings. Our results show that weight-only statistical scoring can match or outperform data-dependent methods for structured head pruning, providing a practical, zero-cost criterion for Transformer compression.
[NLP-32] Differentiable Fuzzy Inference Layer: A Monotone Compositional Ordinal Reasoning Head for Large Language Models
【速读】： 该论文旨在解决大语言模型在处理具有序数关系的量词复合表达（如“most of most students passed”）时出现的语义理解偏差问题。传统分类器头将序数类别视为独立标签，缺乏对数值顺序关系的建模能力以及对量词进行代数组合的机制，导致模型输出与实际比例逻辑不符（如错误地输出“most”而非更合理的“some”）。其解决方案的关键在于提出可微分模糊推理层（Differentiable Fuzzy Inference Layer, DFIL），该结构采用双路径设计：一条为标准分类器分支，另一条为基于有序隶属函数库的标量瓶颈分支。DFIL通过引入单调性约束和基于t-范数（t-norm）的组合运算，实现了对序数量词的结构性组合推理，无需额外的组合训练数据即可完成语义复合。此外，标量分支提供了可解释的残差分析接口，增强了模型决策过程的透明性。该方法在多种大语言模型家族的序数自然语言任务中均展现出显著性能提升。
链接: https://arxiv.org/abs/2609.26113
作者: Zhen Zhang,Amr Alanwar
机构: Technical University of Munich (慕尼黑工业大学)
类目: Computation and Language (cs.CL)
备注:
Abstract:A state-of-the-art language model asked to interpret “most of most students passed” typically answers “most,” though composing two instances of “most” yields a proportion closer to “some.” We trace this failure to an architectural choice rather than a data deficit: standard classifier heads treat ordinal categories as independent labels, with no mechanism to respect their natural ordering or compose them algebraically. We introduce the Differentiable Fuzzy Inference Layer (DFIL), a dual-path prediction head pairing a standard classifier with a scalar-bottlenecked branch grounded in a bank of ordered membership functions. DFIL supplies two structural primitives that a label-only head cannot inherit: monotonicity in the underlying quantity, and compositional reasoning via t-norm operations without any compositional training data. The scalar branch additionally provides an interpretable interface for analyzing residual errors. We instantiate DFIL on ordinal natural-language tasks across diverse LLM families.
[NLP-33] SS: Target-Side Sparsification for Speculative Decoding in Domain-Specific Large Language Models
【速读】： 该论文旨在解决大语言模型推理中，传统推测解码（speculative decoding）方法在目标验证模型（target verifier）上过度依赖全深度密集计算所导致的效率瓶颈问题。现有方法主要聚焦于优化轻量级草稿模型（draft model），而保持目标模型结构不变，但在特定领域推理任务中，这种全深度验证并非最优选择。研究发现，有选择性地跳过目标模型的部分层可降低验证开销，同时提升草稿生成内容的接受率，并维持甚至改善下游任务性能。为此，论文提出TSS（Target-side Sparsification Strategy），一种面向目标模型侧的稀疏化框架。其核心创新在于采用兼顾接受率与评估指标的广度搜索策略，动态探索多层级跳过配置，避免固定优先级的约束；通过构建“领域-配置”映射表并由轻量级跳过控制器调度，实现单个完整目标模型支持多种稀疏验证路径，无需重新训练或永久性参数剪枝。实验表明，在Spec-Bench多个领域、模型规模及推测解码方法下，TSS均显著提升草稿接受率和下游任务表现；在翻译场景中，平均接受长度从2.70提升至4.53（+67.8%），BLEU得分由0.131增至0.237（+80.9%），端到端吞吐量从75.6提升至127.3 tokens/s，实现1.68倍加速。
链接: https://arxiv.org/abs/2609.26100
作者: Haibo Hu,Lianming Huang,Qiao Li,Nan Guan,Chun Jason Xue
机构: City University of Hong Kong(香港城市大学); Mohamed bin Zayed University of Artificial Intelligence(穆罕默德·本·扎耶德人工智能大学)
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
备注:
Abstract:Speculative decoding accelerates large language model inference through collaboration between a lightweight draft model and a target verifier. Existing methods mainly improve the draft side, while the target model is typically kept dense and unchanged. We show that, under domain-specific inference, full-depth target verification is not always the optimal choice. Counter-intuitively, skipping selected target layers can reduce verification cost while simultaneously increasing draft acceptance and preserving, or even improving, downstream task performance. Based on this observation, we propose TSS, a target-side sparsification framework for speculative decoding. TSS employs an acceptance- and metric-aware breadth search to explore multi-layer skip configurations without imposing a fixed priority between the two objectives. The selected configurations are stored in a domain-to-configuration mapping and applied by a lightweight skip controller, allowing one complete target model to support multiple sparse verification paths without retraining or permanent parameter pruning. Experiments on Spec-Bench across multiple domains, model scales, and speculative decoding methods show consistent improvements in draft acceptance and downstream task performance. In Translation setting, TSS increases the average accept length from 2.70 to 4.53 (+67.8%), improves BLEU from 0.131 to 0.237 (+80.9%), and raises end-to-end throughput from 75.6 to 127.3 tokens/s, corresponding to a 1.68X speedup.
[NLP-34] One Domain Many Tongues: Composing Domain and Language LoRAs for Cross-Lingual Remote-Sensing MLLM s without Paired Data
【速读】： 该论文旨在解决遥感（Remote-sensing, RS）多模态大语言模型（Multimodal Large Language Models, MLLMs）在多语言支持方面的瓶颈问题：尽管文本指令数据覆盖超过100种语言，但现有RS MLLM的训练与评估仅限于英文，缺乏有效的多语言扩展方法。其核心解决方案是提出一种名为MODL（Mutually Orthogonal Domain-Language composition）的新训练范式，关键在于通过在每一层引入联合损失项，强制域（领域）LoRA（基于英文遥感图像训练）与语言LoRA（仅基于纯文本训练）更新方向相互正交，从而实现跨模态与跨语言能力的解耦学习。这一正交约束是MODL的核心机制——它有效防止了模型在多语言迁移过程中对原始多语言文本能力的破坏，并避免了因初始化随机性导致的性能退化。实验表明，相较于16种替代方案（包括免训练融合、已有正交变体等），MODL在不使用任何多语言遥感样本的情况下，实现了56%-71%的正确目标语言回答率（显著优于最佳替代方案的27%），同时完整保留了基座模型的多语言文本能力，且在西班牙语任务上超越Qwen2.5-VL-7B。该方法验证了正交约束在跨模态多语言迁移中的关键作用，为零样本多语言遥感理解提供了新路径，但非拉丁文字仍属开放挑战。
链接: https://arxiv.org/abs/2609.26097
作者: Xuechen Li
机构: University of Minnesota, Twin Cities (明尼苏达大学双城分校)
类目: Computation and Language (cs.CL); Computer Vision and Pattern Recognition (cs.CV); Machine Learning (cs.LG)
备注:
Abstract:Remote-sensing (RS) multimodal large language models (MLLMs) are trained and evaluated only in English, while text-only instruction data covers over 100 languages. We propose MODL (Mutually Orthogonal Domain-Language composition), a recipe that adds new languages to an English RS MLLM without a single multilingual RS example: a domain LoRA trained on English RS imagery and a language LoRA trained on text alone are learned jointly, under one loss term that keeps the two updates mutually orthogonal at every layer throughout training. This constraint is the recipe’s active ingredient. Without it, the same training answers RS questions correctly but in English, erases much of the base model’s multilingual text ability, and diverges on one seed in three; sixteen alternatives, from training-free merging to prior orthogonality variants, fail the same way. MODL repairs every failure on every seed: answers are correct and in the target language 56-71% of the time, where the best alternative reaches 27% and most stay below 8%, text ability stays at the level of the untrained base, and on Spanish it surpasses Qwen2.5-VL-7B, with zero multilingual-multimodal data. A single five-language adapter retains English, Spanish, and Vietnamese at full strength across three seeds; non-Latin scripts remain an open boundary.
[NLP-35] SpecialEduBench: Benchmarking Vision-Language Models on Knowledge Skill and Attitude in Language Intervention for Autistic Children
【速读】： 该论文旨在解决自闭症儿童早期干预中，教学行为评估缺乏有效基准的问题。传统评估侧重于模型“知识”的掌握程度，而忽视了其在真实干预场景中“教学表现”的动态适配性，尤其忽略了教师回应需基于儿童即时行为、情境依赖性强且依赖非语言线索（如注视转移、等待时长等）的复杂性。解决方案的关键在于构建首个针对特殊教育领域的多维度评估基准——SpecialEduBench，该基准从知识（4,537项）、技能（200项）与态度（68项）三个维度综合评价模型的教育胜任力，其中态度维度通过192个响应单元整合压力与监控双重情境，充分反映教学决策的复杂性。研究基于真实干预录像数据构建评分体系，由七位特殊教育专家参与命题、评分与审核，并据此优化评判模型的指令以对齐专家参考标准。实验发现，当前前沿视觉-语言模型在事实性知识任务上表现尚可，但在情境化教学任务中差异显著，尤其在高压力情境下失败集中，表明现有模型尚未具备真正的教学适应能力。该基准旨在作为部署前的审计工具，并为面向特殊教育领域定制的生成式教学模型提供研发起点。
链接: https://arxiv.org/abs/2609.26090
作者: Jihoi Na,Taeyeong Kim,Sungjune Kong,Jaemin Jung,Min Joung Park,Kyungtae Joo,Ahhyun Kim,Shim Jaechang,Sooyoung Joo,Dongjin Ka,SeJoong Kim,Jimin Kim,HyunJin Jung,Unggi Lee
机构: 未知
类目: Computation and Language (cs.CL)
备注:
Abstract:Language is the target of most early intervention for autistic children. Because the goal and the method change from child to child, the work falls to a teacher who takes one child at a time and judges each scene as it unfolds. Artificial intelligence is now being brought to that work, yet the benchmarks that reach special education ask what a model knows rather than what it does in front of a child. Building one is not straightforward, since whether a response is good teaching depends on what the child has just done, so no answer key applies. The evidence that settles it is visual as much as verbal, since the length of a wait, a shift of gaze, and the child’s uptake leave no trace in a transcript. We introduce \emphSpecialEduBench, which measures pedagogical competence along knowledge, skill, and attitude, with 4,537 knowledge items and with 200 skill items and 68 attitude items built on recorded intervention, the attitude items crossing pressure with monitoring into 192 response cells. Seven special-education experts wrote, scored, and reviewed the items, and we revised the judge model’s instruction against the reference scores they set. Across eight frontier vision-language models no axis is saturated, since the strongest still fails about a tenth of the honesty cells. The models converge where the knowledge is factual and separate where the task is situated, and the failures gather where pressure is applied. We intend the benchmark as an audit to run before deployment and as a starting point for models built for this domain.
[NLP-36] opoCompress: Topology Aware Token Compression Algorithm for Distributed Edge MoE Inference
【速读】： 该论文旨在解决在资源受限的边缘服务器环境中部署混合专家（Mixture-of-Experts, MoE）模型时，因专家分布于异构服务器而导致的高跨服务器通信开销问题。现有方案分别优化令牌流量或语义压缩，但忽视了拓扑相关的路由成本，导致通信效率与资源利用率不匹配。其解决方案的关键在于提出一种拓扑感知的令牌压缩框架——TopoCompress，通过联合优化令牌压缩、专家部署/复制策略、GPU-CPU资源分配以及协同路由机制，在跨服务器传输量、推理质量与资源使用之间实现平衡。为解耦令牌级压缩与周期级部署之间的耦合关系，TopoCompress采用双时间尺度交替优化：在线快速循环中识别并压缩低重要性且高路由成本的令牌，同时联合调度剩余专家激活；离线慢速循环中根据在线推理累积的压缩后流量动态更新专家部署、复制及资源分配策略。理论分析证明了该方法的可行性、最优性、收敛性与计算复杂度可控性。仿真结果表明，TopoCompress能显著降低跨服务器通信量与部署资源消耗，同时保持可调控的推理质量，从而实现带宽与资源受限边缘环境下高效分布式MoE推理。
链接: https://arxiv.org/abs/2609.26061
作者: Ning Li,Xinyu Wang,Xin Yuan,Wenchao Xu,Athanasios V. Vasilakos,Song Guo,Haijun Zhang
机构: 未知
类目: Networking and Internet Architecture (cs.NI); Computation and Language (cs.CL)
备注: 15 pages, 9 figures
Abstract:Mixture-of-experts (MoE) models improve capacity with moderate overhead by sparsely activating experts per token. However, deploying MoE across resource-constrained edge servers incurs substantial cross-server communication as experts are distributed across heterogeneous servers. Existing placement methods optimize for raw token traffic, while conventional compression considers semantics but ignores topology-dependent routing costs. Consequently, independent optimization leads to inefficient communication and resource utilization. This paper proposes TopoCompress, a deployment- and topology-aware token compression framework for communication-efficient distributed edge MoE inference. It jointly optimizes token compression, expert deployment/replication, GPU-CPU residency, and collaborative routing to balance cross-server transmission, quality, and resource use. To address the coupling between token-level compression and epoch-level deployment, TopoCompress employs a two-timescale alternating optimization. In the online fast loop, it identifies and compresses low-importance, high-routing-cost tokens and jointly routes surviving expert activations. In the offline slow loop, it updates expert placement, replication, and GPU-CPU residency according to post-compression traffic accumulated during online inference. We establish the feasibility, optimality, convergence, and computational complexity. Simulations demonstrate that TopoCompress effectively reduces cross-server traffic and deployment resource consumption while maintaining controllable inference quality, enabling efficient distributed MoE inference over bandwidth- and resource-constrained edge infrastructures.
[NLP-37] Optimizing Denoising Trajectories in dLLM s: A Lightweight Evolutionary Heuristic Approach
【速读】： 该论文旨在解决扩散型大语言模型（Diffusion Large Language Models, dLLMs）在推理阶段依赖于精心设计的去噪调度器（denoising scheduler）却缺乏通用性的问题。现有基于置信度的启发式调度器虽表现良好，但存在“提前结束溢出”（EOS Overflow）和“邻近偏差”（Proximal Bias）两大失效模式，其根源在于Transformer注意力机制中部分位置对无效标记（如[EOS]、[MASK]）赋予过高的注意力权重，从而产生误导性的置信度信号。针对此问题，论文提出一种轻量级进化启发式调度器，利用协方差矩阵自适应进化策略（CMA-ES）优化，动态融合多类启发式特征与上下文均场嵌入（contextual mean-field embedding），仅需393个可训练参数即可实现高度情境敏感的去噪轨迹规划。实验表明，该方法在LLaDA和Dream模型上于四个推理与规划基准测试中持续优于主流基线，包括传统启发式方法、块自回归方法及最新先进方法，是目前参数效率最高的神经调度器。
链接: https://arxiv.org/abs/2609.26052
作者: Zijian Zhao,Dian Jin,Xialiang Tong,Sen Li,Mingxuan Yuan
机构: 未知
类目: Computation and Language (cs.CL); Machine Learning (cs.LG)
备注:
Abstract:Diffusion Large Language Models (dLLMs) have recently emerged as a promising alternative to conventional Auto-Regressive (AR) Large Language Models (LLMs). By leveraging bidirectional attention and parallel decoding, dLLMs enable more efficient generation. However, they require a carefully designed denoising scheduler at inference time (absent during training) whose choice significantly impacts generation quality. While confidence-based heuristic schedulers have shown strong empirical performance, they suffer from two critical failure modes: EOS Overflow and Proximal Bias. Through in-depth analysis of the Transformer’s attention patterns, we reveal that these failures stem from certain positions assigning disproportionately high attention weights to invalid tokens (e.g., [MASK] and [EOS]), which produce misleading confidence signals. Building on this insight, empirical evidence shows that valid attention scores can provide complementary guidance to conventional confidence-based heuristics, yet no single metric consistently excels across all scenarios, implying that the optimal denoising trajectory is highly context-dependent. To address this problem, we propose a lightweight evolutionary heuristic scheduler optimized using the Covariance Matrix Adaptation Evolution Strategy (CMA-ES). Our scheduler dynamically integrates multiple heuristic features with a contextual mean-field embedding, while requiring only 393 trainable parameters. Evaluated on LLaDA and Dream across four reasoning and planning benchmarks, our method consistently outperforms strong baselines, including conventional heuristics, block auto-regressive methods, and recent State-Of-The-Art (SOTA) approaches. To the best of our knowledge, it represents the most parameter-efficient neural scheduler to date. Our code is available at this https URL .
[NLP-38] FIRE: Failure-Informed Runtime Engineering for Reliable Language-Model Agents
【速读】： 该论文旨在解决语言模型智能体在生成有效解决方案后，无法持续稳定交付结果的问题，即“求解成功但执行失败”的可靠性瓶颈。其核心解决方案是引入运行时策略（runtime policies）——一种在预设失败前状态施加针对性自然语言指令或动作拒绝的轻量级机制，无需修改模型权重或用户提示。该方法通过在关键状态施加干预，显著提升方案的重复可交付性。实验表明，在完整的87任务Terminal-Bench 2.1基准上，三类GPT-5.6模型的重复成功（pass^2）率均实现显著提升，其中Sol模型的重复成功率从64.4%提升至73.6%，而单次成功率仅提升1.2个百分点，说明策略主要作用于将“可达解”转化为“可复现交付”。进一步的五臂随机对照实验验证了真实策略的有效性：在可适用任务上达到61%的重复成功率，远高于无策略（39%）、时间匹配伪策略（36%）及通用验证/重思策略（39–43%）。编码分析显示，真实策略在22次干预中均体现预期纠正行为，显著优于其他组别。因此，运行时策略构成了一种高效、可工程化的可靠性增强层，能够在不改变模型能力的前提下，大幅提升智能体现有能力的重复可交付性。
链接: https://arxiv.org/abs/2609.26048
作者: Nikita Agarwal,Nivedit Jain
机构: Failproof AI
类目: Artificial Intelligence (cs.AI); Computation and Language (cs.CL); Software Engineering (cs.SE)
备注:
Abstract:Language-model agents often reach a working solution and then fail to consistently deliver it. We study runtime policies: targeted natural-language instructions and action denials applied by the agent harness at states that preceded observed failures, without changing model weights or the user prompt. With this, keeping capability constant, we observe a meaningful unlock in delivered reliability. Across the complete 87-task Terminal-Bench 2.1 suite, with two attempts per task, policies increase repeated success (pass^2) in all three GPT-5.6 tiers: 50.6% to 54.0% for Luna, 55.2% to 60.9% for Terra, and 64.4% to 73.6% for Sol. Sol’s best-of-two success changes by 1.2 points while repeated success rises by 9.2, showing that policies chiefly convert reachable solutions into dependable delivery. We further cover 14 tasks under Terra’s frozen portfolio. Policy-guided Terra reaches 71.4%, compared with 64.3% for unassisted Sol, at about half the cost, demonstrating how engineering around models could unlock dependability for a use case. To isolate the mechanism we run a randomized five-arm experiment: real policies reach 61% on eligible tasks, versus 39% without a policy, 36% with a timing-matched sham, and 39 to 43% with generic verification or reconsideration. The intended corrective behavior appears in 22 of 24 coded policy attempts, against at most 14 in any other arm. Runtime policies are therefore a practical reliability layer: they make capabilities an agent already possesses substantially more repeatable.
[NLP-39] ruth for Believable AI: Expressed Doubt Provenance and Belief Revision as an Engineerable Stance
【速读】： 该论文旨在解决对话智能体在生成回答时普遍表现出过度自信（overconfident）的问题，即其输出缺乏对自身知识边界的认知、无法有效表达不确定性、缺乏可追溯的证据来源以及难以实现信念修正。为此，作者提出了一种可叠加于固定语言模型之上的行为层（behavior layer）解决方案，其核心在于构建一个包含三种认知状态的机制：基于每条声明的置信度与类型化证据来源（provenance）、基于证据来源的表达触发规则，以及一个具备可审计性记录和部分抗错误修正能力的持久化信念修订存储器。该方案的关键创新在于通过“一致性门控”（consistency-gating）机制实现对表达内容的可信筛选，实验表明以采样一致性为依据的门控策略能显著优于仅依赖平均答案词概率的策略（AUC 0.66 vs. 0.41），并满足对话层面的操作可控性与能力等价性标准，但其有效性具有选择依赖性且在大规模事实集上仍需进一步验证。
链接: https://arxiv.org/abs/2609.26035
作者: Sebastian Cochinescu
机构: University of Bucharest(布加勒斯特大学)
类目: Computation and Language (cs.CL)
备注: 17 pages, 4 figures, 3 tables. Companion framework paper: arXiv:2607.15883 . Code, benchmark, cached model outputs, and result files archived at doi: https://doi.org/10.5281/zenodo.21462986 (code and results) and doi: https://doi.org/10.5281/zenodo.21462988 (benchmark dataset)
Abstract:Conversational agents often express answers in a uniformly confident register. We test whether expressed uncertainty, provenance-aware assertion, and explicit belief revision can be implemented as a behavior layer over a fixed language model; we do not test believability or trust. The layer combines three epistemic states, per-claim confidence and typed provenance, a provenance-gated expression rule, and a persistent revision store with auditable acknowledgments and partial resistance to false corrections. We evaluate it on a constructed, mechanically scored multi-session benchmark using a synthetic model and Qwen2.5-0.5B-Instruct. The synthetic instrument passes all five checks. On the real model, acknowledgment soundness, a by-construction guarantee, holds in 100% of cases, and true corrections are accepted more often than false ones (0.44 vs. 0.15 on held beliefs; 0.875 vs. 0.420 including rule-accepted corrections of unheld facts), but the pre-specified expression-fidelity, contradiction-separation, and provenance margins fail. A disclosed post hoc analysis shows that expression gated on mean answer-token probability ranks correctness below chance end to end (AUC 0.41, conversation-clustered), whereas gating on sampling consistency discriminates (AUC 0.66). A consistency-gated configuration selected from this finding and evaluated under a separately committed protocol meets the conversation-level manipulation and capability-equivalence criteria and replicates on a redrawn conversation set. The manipulation result is selection-dependent, and both criteria remain unresolved when uncertainty is clustered over the 60 facts. The supported conclusions are limited to the by-construction audit guarantee, store-dependent partial correction discrimination, and a benchmark- and model-specific failure of token-probability gating; scaling the fact base is required before human evaluation.
[NLP-40] Domain-Adaptive Pretraining Enhances Water Treatment Semantic Representation for Large-Scale Structured Literature Mining
【速读】： 该论文旨在解决水处理研究领域中知识碎片化严重、缺乏专用语言模型以高效捕捉领域语义和实现大规模文献挖掘的问题。其核心解决方案是提出WaterBERT，一种基于约29.7亿词元的水处理领域大规模语料库进行持续预训练的领域自适应编码器模型。WaterBERT通过在下游任务（如多类处理工艺分类、命名实体识别和关系抽取）上的系统评估，展现出优于通用及特定领域BERT模型的性能，分别达到90.12%、79.50%和74.04%的F1分数。此外，该研究进一步验证了WaterBERT在大规模文献处理中的优势：结合WaterBERTopic方法对5,144篇《环境科学与技术》文章进行主题建模，无需预设类别即可识别出连贯、多样且具有领域特异性的研究主题；基于WaterBERT构建的结构化知识图谱，在处理693,211篇摘要时成本远低于商业大模型，同时保持了竞争力的信息提取性能。最终，将知识图谱与词汇和稠密检索相结合，开发出水知识增强型检索系统（WaterKERS），其相关性得分高达77.7，显著优于传统文本检索基线（54.7–64.5）。因此，WaterBERT为水处理研究中的大规模信息处理与证据映射提供了紧凑且可扩展的语义基础。
链接: https://arxiv.org/abs/2609.26034
作者: Mudi Zhai(1),Ruihong Qiu(2),Qingyun Zeng(3,4),T. David Waite(1),Bing-Jie Ni(1),Haoran Duan(1,5) ((1) UNSW Water Research Centre, School of Civil and Environmental Engineering, The University of New South Wales, Sydney, NSW 2052, Australia (2) School of Electrical Engineering and Computer Science, The University of Queensland, Brisbane, QLD 4072, Australia (3) Microsoft Copilot Studio AI, Redmond, WA 98052, United States (4) Departments of Mathematics amp; Department of Computer and Information Science, University of Pennsylvania, Philadelphia, PA 19104, United States (5) Department of Civil Engineering, The University of Hong Kong, Pokfulam, Hong Kong SAR, China)
机构: UNSW Water Research Centre, School of Civil and Environmental Engineering, The University of New South Wales, Sydney, NSW 2052, Australia; School of Electrical Engineering and Computer Science, The University of Queensland, Brisbane, QLD 4072, Australia; Microsoft Copilot Studio AI, Redmond, WA 98052, United States; Departments of Mathematics Department of Computer and Information Science, University of Pennsylvania, Philadelphia, PA 19104, United States; Department of Civil Engineering, The University of Hong Kong, Pokfulam, Hong Kong SAR, China
类目: Computation and Language (cs.CL)
备注:
Abstract:Water treatment research is expanding rapidly, but much of the knowledge acquired from this research remains scattered across unstructured literature. The field still lacks a dedicated language model that can efficiently capture water treatment-specific domain semantics for large-scale literature mining. Here, we address this by developing WaterBERT, a domain-adapted encoder model designed for semantic representation and structured information extraction from water treatment texts. WaterBERT was developed by continual pretraining on a large-scale water treatment corpus comprising about 2.97 billion tokens. Three fine-tuned models based on WaterBERT were systematically evaluated on downstream tasks, achieving the best overall performance among general-purpose and domain-specific BERT models, with F1 scores of 90.12% for multiclass treatment process classification, 79.50% for named entity recognition, and 74.04% for relation extraction. Beyond these benchmark tasks, we further demonstrated WaterBERT’s advantages for large-scale literature processing. Applied to 5,144 Environmental Science Technology articles, WaterBERT-BERTopic identified coherent, diverse, and domain-specific research topics without predefined categories. Building on WaterBERT, we processed 693,211 abstracts at substantially lower cost than commercial LLMs while retaining competitive extraction performance to construct a structured water treatment knowledge graph. The knowledge graph was then integrated with lexical and dense retrieval to develop a Water Knowledge-Enhanced Retrieval System (WaterKERS), which achieved a relevance score of 77.7, substantially outperforming text-based retrieval baselines (54.7-64.5). Through WaterBERT, this study provides a compact and scalable semantic foundation for large-scale information processing and evidence mapping in water treatment research.
[NLP-41] MICRO: Multi-Fidelity Active Search for Severe Error Discovery ICASSP2027
【速读】： 该论文旨在解决在人工反馈（human feedback）成本与信息量之间存在权衡的场景下，如何高效识别严重错误（severe errors）的问题。具体而言，高成本但信息丰富的强反馈（strong feedback）虽能有效揭示严重错误，但难以大规模使用；而低成本的质量评分（quality ratings）虽可降低标注开销，但其信息量有限。为优化资源分配，论文提出一种名为MICRO（Multi-Fidelity Impact Clustered Rollout）的主动搜索框架，其核心在于通过联合建模不同质量等级反馈（如评分与标注损失）对项目特征的条件依赖关系，动态指导反馈采集策略。MICRO的关键创新在于：首先基于预测的严重性概率对候选样本进行聚类，以确保采样多样性；随后采用滚动估算（rollout）方法评估各候选项在给定预算下的潜在发现价值，从而实现对共享预算的最优分配。实验结果表明，在WMT20英德语翻译数据集上，质量评分不仅能提升损失重构与严重性预测性能，且MICRO在四种预算与评分成本组合下均实现了最高的平均严重错误发现数量，相较于六种基准策略（包括两种滚动控制基线）表现出显著优势（p < 0.001），验证了其在多保真度反馈协同优化中的有效性。
链接: https://arxiv.org/abs/2609.26025
作者: Orlando Leone,Niclas Pokel,Pehuén Moure,Yingqiang Gao,Roman Boehringer
机构: 未知
类目: Machine Learning (cs.LG); Computation and Language (cs.CL)
备注: Submitted to IEEE ICASSP 2027
Abstract:Human feedback can vary in cost and informativeness. Strong feedback can reveal severe errors but is costly, so cheaper quality ratings can help decide which items to annotate. We propose MICRO (Multi-Fidelity Impact Clustered Rollout), an active search framework that allocates a shared budget to these feedback types to maximise confirmed severe error discoveries. MICRO jointly models ratings and annotation losses conditional on item features to steer acquisition. It clusters acquisitions by their predicted impact on severity probabilities to select diverse candidates, then uses rollout to estimate their discovery value. Experiments on WMT20 English-German show that ratings improve both loss reconstruction and severity prediction. MICRO achieves the highest mean discovery count across four budget and rating cost settings, with similar performance to adapted MF-ENS in one and significant gains over all six comparison policies, including two rollout controls, in the other three (p.001) .
[NLP-42] Challenges of Multi-Speaker Extraction for Real Conversational Speech Enhancement
【速读】： 该论文旨在解决在真实多说话人对话场景中，目标说话人语音提取（target-speaker extraction）与多说话人语音提取（multi-speaker extraction）技术面临的两大挑战：一是训练数据中存在过多静音时段，导致模型对有效语音信息的捕捉能力下降；二是注册语音样本（enrolment speech）与实际对话中的目标语音之间存在显著差异，影响模型的泛化性能。针对上述问题，论文提出了一种新的损失函数，通过抑制训练过程中静音段的负面影响，有效提升了语音可懂度指数（STOI）从0.55提升至0.60，以及频带加权分段信噪比（frequency-weighted segmental SNR）从4.35提升至5.12。该解决方案的关键在于设计一种能够自适应地聚焦于语音活动区域、降低静音干扰的损失机制，同时为后续研究注册语音与目标语音不匹配问题提供了量化分析框架。
链接: https://arxiv.org/abs/2609.25948
作者: Robert Sutherland,Stefan Goetze,Jon Barker
机构: 未知
类目: ound (cs.SD); Computation and Language (cs.CL)
备注: Accepted to the International Workshop on Acoustic Signal Enhancement (IWAENC), Cremona, Italy, September 2026
Abstract:Target-speaker and multi-speaker extraction are techniques for extracting speech from a desired speaker or desired speakers in the presence of other speakers and/or noise. Neural network approaches for this task are often trained and evaluated using simulated datasets, with balanced amounts of target speech and speaker enrolment samples which closely match the target speech. However, in real multi-party conversations, participants are often silent for more time than they are speaking, and their enrolment speech samples can differ substantially from the target speech in the conversation. These factors can impact the training and evaluation of these techniques on recordings of real conversations. This work proposes a new loss function, which helps mitigate the effect of excess silence in training, improving STOI from 0.55 to 0.60, and frequency-weighted segmental SNR from 4.35 to 5.12. Additionally, the impact of the mismatch between the enrolment speech and target speech is explored.
[NLP-43] ClusterFewshot: Improving Few-shot Optimization for LLM s workflow
【速读】： 该论文旨在解决大语言模型（Large Language Model, LLM）工作流中少样本演示（few-shot demonstration）选择效率与效果不足的问题。现有方法虽通过引入成功的推理路径来增强提示（prompt），但其演示选择仍依赖随机采样或基于度量的排序，未能充分考虑任务本身的语义结构，导致所选示例代表性不足且优化成本较高。本文提出ClusterFewshot策略，其核心在于将任务的语义结构分析与效用感知评分相结合，通过聚类方式构建具有代表性的演示集合，从而提升演示质量与泛化能力。在基于DSPy的流水线中验证表明，该方法显著降低了优化成本，同时在独立提示调优及混合提示-权重优化场景下均优于以往基于自举（bootstrap）的方法，实现了更高的准确率。
链接: https://arxiv.org/abs/2609.25939
作者: Omri Bar Haim,Shahar Katz,Lior Wolf
机构: Blavatnik School of Computer Science, Tel Aviv University(特拉维夫大学计算机科学学院)
类目: Computation and Language (cs.CL)
备注:
Abstract:The performance of large language model (LLM) workflows often depends on selecting a small set of in-context demonstrations to guide model behavior on new tasks. Recent methods improve this process by augmenting prompts with successful reasoning paths. However, their demonstration selection relies on random sampling or metric-based rankings, overlooking the semantic structure of the task. We propose ClusterFewshot, a strategy that combines semantic structuring with utility-aware scoring to construct representative and effective few-shot demonstration sets. Evaluated within DSPy-based pipelines, ClusterFewshot substantially reduces optimization cost across multiple benchmarks, while consistently improving accuracy relative to prior bootstrap-based methods in both standalone prompt tuning and hybrid prompt-weight optimization.
[NLP-44] Certified Against Which Oracle? Execution Labels Set the Reported Risk of Conformal Abstention for Text-to-SQL
【速读】： 该论文旨在解决文本到SQL（text-to-SQL）任务中生成式AI模型所依赖的置信度评估机制在真实性与可靠性上的根本性问题。现有方法通常依赖于执行一致性（execution consistency）构建不确定性管道，其置信度评分基于单一数据库基准（benchmark）的正确性标签，而该基准所用的数据库被证实具有宽松的“黄金标准”（oracle）特性，导致置信度评估存在系统性偏差。研究通过预注册干预，在Spider-Realistic数据集上将原基准数据库替换为经过提炼的多实例测试套件（distilled multi-instance test suite），发现该替换使置信证书在保留样本上的风险（risk）相比其自身标签报告值上升了2.73至10.23个百分点，显著暴露了原有评估体系的不准确性。进一步分析表明，无论是宽松还是更严格的黄金标准，均未能反映人工专家标注的真实风险水平：在双专家盲标条件下，一个名义置信度为0.10的证书实际风险分别高达20.0和17.2点。更严重的是，严格黄金标准在判断上出现双重错误——既误拒了本应正确的答案，也误接受了部分错误答案。对拒绝样本的AI普查显示，其中约四分之一至三分之一存在语义错误，其余主要归因于问题描述不明确、合成实例或疑似参考查询缺陷，该结论得到预注册盲审专家审计的支持。此外，研究揭示了置信度评分的评估结果高度依赖于其训练所用的黄金标准：在16种组合中，任何执行一致性得分在其所属黄金标准的标签下均表现更优；而在专家标签下，若使用测试套件聚类而非原始数据库聚类构建得分，则其受试者工作特征曲线下面积（AUROC）提升6.96和1.53个百分点，且后者的专家区间完全排除了套件标签报告的8.3点增益。因此，该研究的关键解决方案在于强调：置信证书应同时报告两种黄金标准下的评估结果，并仅在基准完成独立审计后，方可将两者差异解释为语义风险；同时，一致性得分必须在未参与其聚类构建的黄金标准下进行评估，以确保评估的独立性和有效性。
链接: https://arxiv.org/abs/2609.25938
作者: Jiamiao Liu,Dewen Qiao,Yu Zhang,Xuetao Chen
机构: Xinqiao Hospital (新桥医院); Army Medical University (第三军医大学); Chongqing 400037, China
类目: Machine Learning (cs.LG); Computation and Language (cs.CL)
备注:
Abstract:A conformal abstention certificate for text-to-SQL is only as truthful as the correctness labels it is calibrated on. The uncertainty pipelines that read confidence off execution consistency take those labels from the single database a benchmark ships, an oracle known to be lenient. We run a preregistered intervention on Spider-Realistic, swapping that database for the benchmark’s distilled multi-instance test suite. Across four SQL-specialist checkpoints and two split schemes, the swap raises the certificate’s held-out risk 2.73 to 10.23 points above the risk its own labels report. Neither oracle reports the risk experts assign. Under blinded labels from two SQL experts, a certificate calibrated at a nominal 0.10 carries 20.0 and 17.2 points of risk on two checkpoints. The stricter oracle errs in both directions: most of the answers it rejects are not judged wrong, and some of those it accepts are. An AI-assigned census of what it rejects finds a semantic error in a quarter to a third of them, depending on the population. It attributes most of the rest to underspecified questions, synthetic instances or suspected reference-query defects, a flag supported by a preregistered blinded expert audit. The oracle also decides how a confidence score is judged. Every execution-consistency score looks better under the labels of the oracle that built its clusters, in 16 of 16 combinations. Under expert labels, building such a score on suite clusters instead of shipped-database clusters raises its area under the ROC curve (AUROC) by 6.96 points on one checkpoint and 1.53 on the other. On the second, the expert interval excludes the 8.3 points the suite labels report. A certificate should be reported with both oracles, and an oracle-relative difference read as semantic risk only after the benchmark is audited. A consistency score should be evaluated under an oracle that did not build it.
[NLP-45] Informed Masking: Structure-Aware Perturbation for Reinforcement Learning in Diffusion Large Language Models EMNLP2026
【速读】： 该论文旨在解决扩散型大语言模型（Diffusion Large Language Models, dLLMs）在基于强化学习（Reinforcement Learning, RL）对齐过程中，由于单次回滚（rollout）中蒙特卡洛采样预算有限，导致似然代理（likelihood surrogate）估计不准确的问题。现有方法采用均匀随机掩码构建重建子问题，但未考虑不同子问题的优化难度差异，从而影响训练效率与稳定性。论文发现dLLM回滚过程中存在系统性的上游/下游结构：某些词元（token）在解码后会引发邻近未解码位置显著的置信度变化，称为上游词元；而另一些词元仅引起局部微小变化，称为下游词元。研究揭示，掩码下游词元比掩码上游词元能生成更良构的子问题，这一现象被称为“子问题难度不对称性”。针对此问题，作者提出有知掩码（Informed Masking, IM），通过分析去噪轨迹（denoising trajectory）在零额外推理成本下为每个词元计算优先级得分，并引导掩码采样偏向下游词元。该方法具有即插即用特性，在三个先进的dLLM强化学习方法上应用于LLaDA-8B-Instruct数据集，分别在数学和规划基准测试中实现最高2.01%、8.68%和5.77%的相对平均性能提升，同时显著改善训练稳定性。其核心创新在于利用模型内部动态信息识别并优先处理更具优化价值的子问题，从而高效提升似然代理质量。
链接: https://arxiv.org/abs/2609.25927
作者: Xiaoyi Yu,Enver Sangineto,Pei Fu,Fiorenzo Parascandolo,Wenhui Tan,Ruikang Zhang,Rita Cucchiara,Ruihua Song,Jian Luan
机构: Gaoling School of Artificial Intelligence, Renmin University of China (中国人民大学高瓴人工智能学院); MiLM Plus, Xiaomi Inc. (小米公司); University of Modena and Reggio Emilia (摩德纳和雷焦艾米利亚大学); Peking University (北京大学)
类目: Computation and Language (cs.CL)
备注: 17 pages, 4 figures, EMNLP2026 Findings
Abstract:Diffusion Large Language Models (dLLMs) have emerged as an efficient alternative to autoregressive models, yet aligning them via Reinforcement Learning (RL) requires likelihood surrogates estimated from masked reconstruction subproblems under a small Monte Carlo budget per rollout. Existing methods construct these subproblems by uniform random masking, leaving open the question of which subproblems to prioritize. We identify a systematic upstream/downstream structure in dLLM rollouts. Some tokens, when revealed, trigger large confidence changes in nearby undecoded positions; we call them upstream. Others induce only small local changes and are therefore downstream. We find masking downstream tokens yields substantially better-posed subproblems than masking upstream tokens, a phenomenon we term subproblem difficulty asymmetry. Based on the observation, we propose Informed Masking (IM), which derives a per-token priority score from the denoising trajectory at zero extra inference cost and biases mask sampling toward downstream tokens. IM is plug-and-play: when plugged into three state-of-the-art dLLM RL methods on LLaDA-8B-Instruct, it delivers up to 2.01%, 8.68%, and 5.77% relative average gains on math and planning benchmarks with improved training stability.
[NLP-46] Rethinking Length-Based Training: Batch Composition and Loss Normalization in Speech Token Language Models
【速读】： 该论文旨在解决短序列到长序列训练（short-to-long training）在语音模型中的有效性问题，尤其是其在生成式语音语言模型中带来的性能提升机制难以解释的挑战。研究表明，在基于批量均值损失（batch-mean loss）的训练框架下，长度相关的训练策略会通过改变打乱策略、批次构成、词元保留率及词元权重等多个因素共同影响模型性能，但这些因素之间存在耦合效应，难以单独评估其贡献。为此，作者通过匹配对比实验对这些混杂因素进行解耦分析。研究发现，当批次构成与词元暴露量固定时，仅采用从短到长的训练顺序并无独立优势；而在首个训练周期中采用分组训练可降低Mimi模型的困惑度（perplexity），但此优势在词元平衡损失（token-balanced loss）下消失。此外，跨分词器（cross-tokenizer）的实验结果支持了词元长度变化与词元权重之间存在关联的假设。该工作提出了一套系统性分析框架，为研究变长语音模型中的长度依赖训练策略提供了方法论基础。
链接: https://arxiv.org/abs/2609.25890
作者: Hongjin Song,Runwu Shi,Weiqiao Shan,Jiale Luo,Yujin Wang,Yifei Wu,Chunxiang Jin
机构: 未知
类目: Computation and Language (cs.CL)
备注:
Abstract:Short-to-long training is a simple curriculum for speech models, but its gains can be difficult to interpret. In speech token language models, length-based training can change the shuffle policy, batch composition, token retention, and token weights under batch-mean loss. We disentangle these factors through matched comparisons. In the tested settings, short-to-long ordering shows no independent benefit when batch composition and token exposure are fixed. First-epoch grouping lowers perplexity for Mimi under batch-mean loss, but this gain is not observed under token-balanced loss. The cross-tokenizer results are consistent with a link between chunk-length variation and token weighting. This work provides a systematic analysis protocol for studying length-based training in variable-length speech models.
[NLP-47] Isolated Sign Language Recognition for Icelandic Sign Language: Experiments in a Low-resource Setting
【速读】： 该论文旨在解决冰岛手语（ÍTM）这一极度低资源语言在孤立手语识别（ISLR）任务中的识别难题。其核心挑战在于数据稀缺：所使用的ÍTM SignWiki数据集仅包含1,845个视频，覆盖849个类别，其中86%的类别仅有两个样本，导致整个任务本质上属于跨签名者的一次性识别（one-shot recognition）。为应对这一挑战，研究提出的关键解决方案包括：采用先进的开源ISLR框架（SPOTER优于OpenHands），结合高效的姿态估计器（MediaPipe表现最优），并引入跨语言迁移学习策略。实验表明，通过在资源更丰富的美国手语（ASL）数据上预训练后微调SPOTER，准确率提升14–24个百分点；而利用六种其他手语的多语言联合训练，使OpenHands在全量任务上的性能从1.41%大幅提升至28.86%。结果表明，从高资源手语向低资源手语进行迁移学习是可行且有效的路径，为未来低资源手语识别系统的发展提供了重要方向。研究还发布了两个框架的适配版本以促进社区共享与复现。
链接: https://arxiv.org/abs/2609.25862
作者: Finnur Ágúst Ingimundarson,Guðný Björk Þorvaldsdóttir,Mathias Müller,Sarah Ebling
机构: University of Zurich (苏黎世大学); Communication Centre for the Deaf and Hard of Hearing in Iceland (冰岛聋人和听力障碍者通讯中心)
类目: Computation and Language (cs.CL)
备注:
Abstract:We present the first experiments on isolated sign language recognition (ISLR) for Icelandic Sign Language (ÍTM). We use ÍTM SignWiki, a dataset derived from a bilingual Icelandic–ÍTM online dictionary. It is genuinely low-resource: 1,845 videos cover 849 classes, 86% of which have only two examples, making the full task effectively one-shot recognition across signers. We compare two open-source ISLR frameworks, OpenHands and SPOTER, on three tasks of increasing vocabulary size (22, 117 and 849 classes), and evaluate three pose estimators and two forms of cross-lingual transfer. With ÍTM data alone, SPOTER outperforms OpenHands on all three tasks, and MediaPipe poses give better results than AlphaPose or SDPose. Cross-lingual transfer brings the largest gains: pretraining SPOTER on American Sign Language data before finetuning on ÍTM raises accuracy by 14–24 percentage points, to 72.7%, 47.9% and 22.6% on the three tasks, and multilingual training with data from six other sign languages lifts OpenHands from 1.41% to 28.86% on the full task. Although far from practical use, the results suggest that transfer from better-resourced sign languages is promising for very low-resource ones. We release our adapted versions of both frameworks.
[NLP-48] BELXTR: Biomedical Entity Linking via Contextualized Token Retrieval
【速读】： 该论文旨在解决生物医学实体链接（Biomedical Entity Linking, BEL）中嵌入式模型因将提及项（mentions）和实体（entities）压缩为单一向量而导致细粒度差异被平均化的问题。其核心解决方案是提出BELXTR，一种基于多向量（multi-vector，又称晚期交互）架构的新型嵌入模型，能够有效利用标记级匹配信息。BELXTR通过引入任务特定的训练目标并探索主动查询扩展策略，在十个语料库和五个知识库上的实验表明，其在半数数据集上超越当前最优方法，平均召回率@1提升5个百分点；尤其在具有挑战性的跨物种基因消歧子任务中表现突出，优于基于大语言模型（LLM）的检索-重排序流水线，并接近专用规则系统性能。结果表明，多向量模型可作为难以维护的规则系统或在大规模文献挖掘场景下成本过高的LLM重排序的实用替代方案。
链接: https://arxiv.org/abs/2609.25859
作者: Samuele Garda,Ulf Leser
机构: 未知
类目: Computation and Language (cs.CL)
备注:
Abstract:Biomedical Entity Linking disambiguates mentions to entities in a knowledge base (KB), making it the cornerstone of information extraction pipelines. While embedding-based models are a popular approach for the task, they suffer from a key limitation. They compress mentions (and entities) into a single vector, forcing the model to average away crucial fine-grained differences. We present BELXTR, a novel embedding model based on the multi-vector (a.k.a. late interaction) architecture, which allows to leverage token-level matching information. BELXTR extends the original XTR model to biomedical entity linking by integrating an existing task-specific training objective and exploring active query expansion. Experiments across ten corpora and five KBs show that BELXTR improves upon current state-of-the-art in half of the corpora with an average improvement of 5pp recall@1. The largest gains are reported on the challenging cross-species gene disambiguation subtask, where BELXTR outperforms an LLM-powered retrieve-and-rerank pipeline and closely approaches a specialized rule-based system. Our results highlight multi-vector models as a practical alternative to hard-to-maintain rule-based systems or in scenarios where LLM-based reranking is too costly as in PubMed-scale mining. The code to reproduce our experiments can be found at: this https URL.
[NLP-49] MemoryAthena: Adaptive Routing over Latent and Generated Memories
【速读】： 该论文旨在解决传统记忆增强型模型中仅依赖显式记忆表的检索（即直接记忆读取，Engram retrieval, E）而无法动态生成上下文相关记忆表示的问题。其核心挑战在于如何在不破坏已有可靠记忆的前提下，有效引入由模型内部状态生成的、具有条件性优势的新记忆表示。解决方案的关键在于提出MemoryAthena框架，该框架通过三条并行路径实现：（1）直接记忆检索（E），（2）基于检索到的记忆线索生成记忆（GE），以及（3）完全脱离记忆表、从因果主干状态直接生成记忆（GH）。其中，生成式记忆（GE/GH）并非始终有益，其有效性依赖于具体任务与输入情境。因此，系统将直接检索（E）作为基准锚点，通过训练一个轻量级因果路由头（causal routing head），学习在何种条件下、以何种强度对E的输出进行干预。该路由头基于反事实未来词元预测优势（counterfactual future-token likelihood advantages）进行优化，在推理时通过有界插值调整E的残差（residual）以实现选择性修正，或在不采纳时完全恢复原始路径。实验表明，该方法在五项问答任务上将平均得分从37.65提升至39.28，在六项通用自然语言任务上从76.73提升至79.13，验证了生成记忆作为选择性修正机制的有效性。研究进一步揭示，不同路径间存在任务与输入层面的互补优势，凸显“何时、何地、以多强程度进行干预”这一路由决策是生成记忆系统的核心挑战。
链接: https://arxiv.org/abs/2609.25853
作者: Mingyuan Li,Guangsheng Yu,Juyuan Zhang,Xu Wang,Zhibo Man,Haonan Zhang,Shaoxiong Ji
机构: ELLIS Institute of Finland(ELLIS芬兰研究所); University of Turku(图尔库大学); University of Technology Sydney(悉尼科技大学); University of Science and Technology of China(中国科学技术大学); Shanghai Jiao Tong University(上海交通大学)
类目: Computation and Language (cs.CL)
备注:
Abstract:Learned-memory methods store information in an explicit table and consume it through a separate reader, allowing addressing, storage, and reading to be modified independently. We study whether useful memory can also be generated rather than only retrieved. MemoryAthena uses three pathways: direct Engram retrieval (E), generation from retrieved Engram cues (GE), and generation from causal backbone states without consulting the memory table (GH). Generated memory is conditionally useful: it can complement E in one context but interfere with it in another. MemoryAthena therefore treats E as an anchor and learns when a generated representation should intervene. With the backbone, memory, generators, and readers frozen, a lightweight causal routing head is trained from counterfactual future-token likelihood advantages of GE and GH relative to E. At inference time, an admitted candidate modifies the E residual through bounded interpolation, while rejection recovers the direct pathway exactly. On question answering, MemoryAthena raises the five-task average from 37.65 to 39.28 over the direct pathway of the same checkpoint, while the six-task general-NLP average increases from 76.73 to 79.13. The complete memory-side system contains approximately 201M parameters, excluding the frozen backbone. Further analyses show complementary strengths among E, GE, and GH across tasks and inputs. These results support generated memory as a selective correction to direct retrieval and highlight routing when, which, and how strongly to intervene as the central challenge.
[NLP-50] Auditing Proxy-Based Validation Across Text Spans
【速读】： 该论文旨在解决评估分数（evaluation score）在依赖廉价代理标签（proxy label）进行验证时，因评估分数与代理标签共享相同文本片段而产生的表面证据偏差问题。这种偏差可能导致评估分数与真实目标构念（target construct，如正确性）之间的相关性被高估，从而误导对模型性能的真实判断。其解决方案的关键在于引入“验证契约”（validation contract）的概念，明确界定评估分数、其对应文本跨度、代理标签及目标构念之间的关系，并严格在评分跨度之外重新评估代理规则的适用性。通过在HotpotQA任务中控制共享文本边界并进行跨跨度验证，研究发现：当仅使用50字符前缀时，评估分数与代理标签的吻合度显著高于与正确性的吻合度（差距达+0.184），但该差距在120字符后降至≤+0.045；此时评估分数仍能预测答案字符串是否出现在后续内容中（AUC=0.634），而等效性检验表明其与正确性的关联处于随机水平。在OR-Bench上，移除模型重复出现的开头模板可大幅削弱评估分数与拒绝代理标签的关联，而匹配体积删除则无此效果，且构念一致性维持在随机水平。仅有三个外部验证契约支持跨跨度控制，且所采样路由研究均未公开生成文本以供重读代理标签。因此，论文呼吁：基于代理标签的验证声明必须明示各标签的读取跨度，报告代理一致性和构念一致性双指标，并公开生成文本以便实现独立复现与验证。
链接: https://arxiv.org/abs/2609.25808
作者: Daein Weon,Dong Ho Kang
机构: Kookmin University (首尔国立大学); UStechlab
类目: Machine Learning (cs.LG); Computation and Language (cs.CL)
备注: 63 pages, 7 figures, 38 tables. Code: this https URL
Abstract:Evaluation scores are often validated by their agreement with inexpensive proxy labels. When the score and the proxy are computed from the same text span, however, that agreement can arise from surface evidence the two share rather than from the semantic construct the proxy is meant to represent. We make the distinction explicit by declaring the score, its span, the proxy and the target construct as a validation contract, then re-evaluating that proxy rule strictly outside the scored span. In a controlled HotpotQA correctness experiment varying only the shared text boundary, the score agrees with its proxy far better than with correctness at a 50-character prefix: the gap is +0.184, collapsing to at most +0.045 from 120 characters onward. At that short prefix the score still predicts whether the answer string appears later (AUC 0.634) while an equivalence test places its agreement with correctness at chance, so the reported proxy agreement does not establish that the score ranks correctness. On OR-Bench, suppressing each model’s recurring opening templates removes most of the score’s association with the refusal proxy, while matched-volume deletion removes almost none and construct agreement stays at chance. Only three of eleven external contracts support the off-span control, and none of the routing studies we sampled released the generations it needs. We therefore ask that a proxy-based validation claim declare the span each label is read from, report the construct agreement beside the proxy agreement, and release the generations that let the proxy be re-read off the scored span.
[NLP-51] Latest Exact Match Attention
【速读】： 该论文旨在解决传统Transformer模型中注意力机制在计算效率与记忆容量之间的权衡问题，特别是针对长序列建模中注意力开销随上下文长度增长而急剧上升的瓶颈。其核心挑战在于如何在保持强表达能力的同时实现高效且可扩展的注意力计算。解决方案的关键是提出一种新型硬注意力机制——最新精确匹配注意力（Latest Exact Match Attention, LEMA），该机制通过将查询（query）和键（key）二值化（binarized），并强制每个查询仅与最新的精确匹配键进行交互，从而实现稀疏、确定性的注意力模式。这一设计不仅在理论上证明了LEMA Transformer能够模拟字随机存取机（word-RAM），且其反向模拟也具备常数时间复杂度（与上下文长度无关），从而建立了与word-RAM之间在计算与内存使用上的紧密对应关系。此外，论文提出一种基于直通估计器（straight-through estimator）处理二值化不可导操作，并结合渐进退火的软注意力代理（soft attention surrogate）进行训练的方法，在合成关联回忆任务中展现出超越固定状态大小的门控DeltaNet（GDN）的存储与召回能力。大规模实验表明，8.34亿参数的LEMA语言模型在损失上媲美约其一半规模的Softmax Transformer，且在长距离依赖任务中表现优于同规模GDN模型，同时通过基于字典的推理机制实现了恒定生成速度，且字典可驻留于主存而非显存，显著提升了实际部署的可扩展性。
链接: https://arxiv.org/abs/2609.25802
作者: Moritz Brösamle
机构: University of Tübingen (图宾根大学)
类目: Machine Learning (cs.LG); Computational Complexity (cs.CC); Computation and Language (cs.CL)
备注:
Abstract:We introduce latest exact match attention (LEMA), an attention variant for transformers where queries and keys are binarized and each query attends only to the latest exactly matching key. We prove that LEMA transformers with chain of thought can simulate word-RAMs, as was recently shown for the less restrictive rightmost hard attention. In contrast to prior hard attention variants, the restriction to exact matches enables an efficient converse direction: word-RAMs can simulate LEMA transformers at a cost per token independent of the context length. Together, these results yield a close correspondence between the two computational models in terms of both compute and memory. Beyond the theory, we propose a training method for LEMA transformers that handles their non-differentiable operations with a straight-through estimator for the binarization and a soft attention surrogate annealed towards LEMA. On a synthetic associative recall task, LEMA models trained this way use their growing state to store and recall a large number of associations, outperforming gated DeltaNet (GDN) with its fixed state size. As a first scaling test, we train LEMA language models with up to 834 million parameters. They match softmax transformers of around half their size in loss and, on repeated rare phrases and a needle-retrieval task, remain behind softmax transformers but recall across longer distances than GDN models of comparable size. Finally, we implement dictionary-based inference for LEMA transformers and show constant generation speed comparable to GDN despite their growing state, with the dictionaries residing in main memory rather than VRAM. Code is available at this https URL.
[NLP-52] Reply to comments arXiv:2512.07881 and arXiv:2601.06104 on quantum structure in human and AI-generated language
【速读】： 该论文旨在回应关于其在人类语言中量子力学统计（arXiv:2407.14924）及生成式人工智能语言中的量子结构（arXiv:2511.21731）研究的批评意见，重点解决以下核心问题：一是大语言模型（LLM）实验中所采用协议的探索性本质；二是边缘律违背（marginal-law violations）与“默认上下文依赖性”（Contextuality-by-Default）准则在识别量子纠缠中的作用；三是单独使用玻色-爱因斯坦（Bose-Einstein）拟合的诊断价值有限性；四是将最低能量层级分配给最频繁词汇的意义；以及大语言模型所用向量空间与量子态空间之间的关联性。其解决方案的关键在于系统性澄清上述争议点，强调方法论的探索性质，并指出仅凭单一拟合模型无法充分表征复杂语义结构，同时阐明语义分布与量子态空间之间可能存在但需进一步验证的映射关系。
链接: https://arxiv.org/abs/2609.25797
作者: Massimiliano Sassoli de Bianchi,Roberto Leporini
机构: Center Leo Apostel for Interdisciplinary Studies, Vrije Universiteit Brussel (VUB)(弗莱堡自由大学中心利奥·阿波斯泰尔跨学科研究所); Department of Economics, University of Bergamo(贝加莫大学经济系)
类目: Computation and Language (cs.CL)
备注: Reply to comments arXiv:2512.07881 and arXiv:2601.06104 , 6 pages
Abstract:We reply to the comments by M. Sienicki and K. Sienicki (arXiv:2512.07881) and by K. Sienicki (arXiv:2601.06104) on our work on quantum-mechanical statistics in human language (arXiv:2407.14924) and on quantum structure in AI-generated language (arXiv:2511.21731). We thank the authors for their careful reading and address what we consider to be the main points of criticism: the exploratory nature of the protocol used in the experiments with large language models; the role of marginal-law violations, and of the Contextuality-by-Default criterion, in the identification of entanglement; the limited diagnostic value of a Bose-Einstein fit taken in isolation; the meaning of assigning the lowest energy levels to the most frequent words; and the relation between the vector spaces used by LLMs and quantum state spaces. We also correct a typographical error in Table 3 of arXiv:2511.21731, which does not affect the reported CHSH value.
[NLP-53] Syndrome Synergy and Safety: Structured Reasoning and Knowledge-Driven Alignment for TCM Prescription Generation
【速读】： 该论文旨在解决大语言模型在中医处方生成应用中面临的三大临床关键缺陷：缺乏可审计的辨证论治推理路径（即“理法方药”范式下的可解释性缺失，简称SR Gap）、单次诊疗孤立处理而无法根据“随证加减”进行动态调整治疗方案（称为LA Gap），以及未能严格遵守如“十八反”等绝对禁忌用药规则（SC Gap）。其解决方案的核心在于提出一个渐进式的四阶段框架（SFT → PG-CoT → Dynamic → K-RL）：通过基于“理法方药”范式的生成式思维链（PG-CoT）对思维链蒸馏过程进行约束，实现可追溯的诊断推理链条；利用动态微调（Dynamic SFT）建模患者病程轨迹并引入显式转移推理机制以支持随证加减；最后采用基于规则的强化学习（K-RL），将确定性的药理禁忌规则编码为基于偏好（DPO）的规则信号，从而确保处方生成符合中药配伍禁忌。该框架在12个微调模型与6个零样本基线对比中显著提升处方质量，其中70亿参数模型（Mistral-7B）在所有三项中医评估指标上均超越零样本GPT-5。
链接: https://arxiv.org/abs/2609.25755
作者: Zheng Chen,ZhiCheng Du,Haoxuan Li,Peiwu Qin
机构: Tsinghua University (清华大学); Guangdong Provincial Laboratory of Traditional Chinese Medicine Hengqin (广东省中医药横琴实验室)
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
备注: 21pages, 6figures
Abstract:Applying large language models to Traditional Chinese Medicine (TCM) prescription generation reveals three clinically critical gaps: models produce end-to-end mappings without auditable reasoning following the li-fa-fang-yao paradigm (SR Gap), treat each encounter in isolation without follow-up adjustment via sui zheng jia jian (LA Gap), and fail to enforce absolute contraindication rules such as Shi Ba Fan (SC Gap). We propose a progressive four-stage framework (SFT \to PG-CoT \to Dynamic \to K-RL) that addresses each gap: PG-CoT constrains CoT distillation under the li-fa-fang-yao paradigm to produce auditable diagnostic chains, Dynamic SFT models patient trajectories with explicit transition reasoning, and K-RL encodes deterministic pharmacological rules as rule-based DPO preference signals. Across 12 fine-tuned models and 6 zero-shot baselines, our framework substantially improves prescription quality over zero-shot baselines—with a 7B model (Mistral-7B) surpassing zero-shot GPT-5 on all three TCM evaluation metrics.
[NLP-54] Slow Decay and Silenced Expression: Iterated Subliminal Trait Transfer in Language-Model Lineages
【速读】： 该论文旨在解决生成式 AI（Generative AI）在多代模型链（lineage）中隐性特质（trait）的传播与持久性问题。随着语言模型越来越多地基于其他模型的输出进行训练，形成具有继承关系的模型世代链，研究其内部特质是否能在跨代传递中保持或衰减，成为关键挑战。其解决方案的关键在于：通过构建多代模型链并采用双重评估机制——即关键词筛查（keyword screen）检测行为层面的特质表达，以及激活探针（activation probe）分析模型内部状态相对于基线的位移方向——来量化特质在代际间的保留情况。研究发现，尽管行为层面的显性表达率随代际递减（从55.6%降至21.1%），但特质在内部表征中仍持续存在；更关键的是，在移除默认系统提示（system prompt）后，即使模型不再表现出任何关键词特征，激活探针仍能检测到正向信号，表明特质已内化于模型参数中。此外，通过将高阶学生模型的位移施加于基线模型，可诱导其在无系统提示下重现该特质，进一步证实了特质的内在可引导性。这揭示出生成式模型链中潜在偏见或控制信号可能以非显性方式长期潜伏并被重新激活，对模型安全与可控性提出深远警示。
链接: https://arxiv.org/abs/2609.25721
作者: Ryan Vo,Duc-Vu Nguyen,Matt Kretchmar,Ngan Luu-Thuy Nguyen
机构: 未知
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Computation and Language (cs.CL)
备注: 7 pages plus appendix. Extended version with additional experiments to follow
Abstract:Language models are increasingly trained on the outputs of other models, forming chains that we call lineages, in which a trait present in one generation can pass to the next. Prior work on subliminal learning has shown that a teacher’s trait can transmit to a student through filtered data carrying none of the trait’s content. However, the evidence covers only a single training step. We study whether such a trait holds or fades across lineages. We instill the trait into three copies of Qwen2.5-7B-Instruct and iterate the training step to depth ten from each, reading every generation two ways on the same held-out prompts: a keyword screen that looks for expressions of the trait in the model’s output, and an activation probe that projects each model’s displacement from the base onto a direction built from the other lineages’ teachers. We report two findings. First, the trait persists through ten generations across three lineages. The instilled models express it on every completion; the keyword-screen rate falls to 55.6% after the first step and to 21.1% by generation ten. The base itself matches the screen on none of its 300 completions. Second, the trait can be present internally while absent behaviorally. When the model’s default system prompt is removed at evaluation, the generation-ten students’ keyword-screen rate is zero on every prompt while the probe score stays positive on every prompt. Steering the untreated base with the displacement of a generation-ten student, which is trained and measured under the default system prompt, induces screened expression of the trait even with the system prompt removed, while that same student shows no expression of the trait with the system prompt removed.
[NLP-55] How Strongly Should Task State Influence an LLM Agent ?
【速读】： 该论文旨在解决长时序任务中大语言模型（LLM）代理对任务状态跟踪的可靠性问题，即如何有效管理任务的执行状态（如已完成、阻塞、取消或可重试），并探究不同状态传递机制对代理表现的影响。其核心挑战在于：现有系统将状态信息以文本形式嵌入提示（prompt）、依赖模型自行解读，或通过独立模块强制执行，但缺乏对状态呈现、告知与强制三种方式各自贡献的量化评估。本文提出一种可控实验设计，固定任务规则、模型和任务实例，仅改变状态信息传递强度：从原始转录文本、精确检查清单、基于状态机生成的逐轮指令，到由状态机驱动的强制门控机制。通过动态真实值进行精确匹配评分，揭示四个关键发现：（1）仅显示准确状态不可靠；（2）代理自写的未验证状态记录优于被展示的精确检查清单；（3）指令有效性与模型服从性正相关；（4）强制机制无需依赖模型服从性，但受限于状态正确性和请求-步骤映射器的准确性。在235B模型上，基于\tau^2-bench航空政策编译的强制门控使通过率从0.39提升至0.54，而35B模型因极少违规未受影响；在PM-Bench任务中，由于行为依赖线索识别而非状态决策，显示记录反而最优，强制机制则导致35B模型性能下降。因此，强制机制在状态决定型且频繁出错的任务中有效，但在判断错误时会带来负面影响。
链接: https://arxiv.org/abs/2609.25686
作者: Chenyu Zhang,Wonbin Kweon,Jiawei Han
机构: University of Waterloo(Waterloo大学); Sungkyunkwan University(成均馆大学); University of Illinois Urbana-Champaign(伊利诺伊大学厄本那-香槟分校)
类目: Artificial Intelligence (cs.AI); Computation and Language (cs.CL); Machine Learning (cs.LG)
备注: Preprint. 43 pages
Abstract:Long-horizon assigned work requires an LLM agent to track the state of a task: which steps are done, blocked, cancelled, or open to repetition. Agent systems either keep this state as text in the prompt and rely on the model to read that text, or move the state into a module that enforces it, and each system is evaluated as a whole, so no one knows how much reliability comes from the state being shown, told, or enforced. We fix the task rules, the model, and paired episodes and vary how strongly task state reaches the agent: a raw transcript, an exact checklist, per-turn directives from a state machine compiled from the brief and advanced only by execution receipts, or an enforcement gate on that machine that refuses state-violating actions; every episode is scored by exact payload matching against dynamic ground truth. Across three models, two reasoning regimes, and two domains, four findings hold without per-turn reasoning: displaying accurate state is unreliable, an unverified ledger the agent writes itself beats an accurate checklist it is shown, directives help in proportion to the model’s obedience, and enforcement needs no obedience but is bounded by the correctness of its state and by the matcher that maps requests to steps; per-turn reasoning at a 235B agent compresses these separations without repairing the text rungs. The same gate, compiled from \tau^2 -bench’s airline policy, raises a 235B agent’s pass ^1 from 0.39 to 0.54 and changes nothing for a 35B agent that rarely violates the policy; on PM-Bench, where acting turns on recognizing a cue rather than on state, showing the record is the best rung–matching or beating both gates and reversing the ledger-over-checklist finding–and enforcing the matcher’s judgement drops a 35B agent below its raw transcript. Enforcement pays when failures are state-decidable and frequent, and hurts when the gate’s judgement is wrong.
[NLP-56] From Utterances to Networks: Modelling Slang Adoption and Diffusion Across Subreddits EMNLP2026
【速读】： 该论文旨在解决在线社区中新词（neologisms）传播机制的复杂性问题，特别是如何在社会互动与语言特征双重维度下理解网络俚语的采纳与扩散过程。传统研究多局限于单一视角——或聚焦于社交互动，或关注俚语自身的语言属性，而未能整合二者。其关键挑战在于大规模在线文本中对俚语使用的标注成本过高。本文创新性地利用大语言模型（LLM）作为可扩展的标注工具，首先构建了一个基于真实Reddit数据的人工标注基准以评估LLM在识别俚语使用上的性能，随后基于LLM生成的标注数据建模俚语的采纳与扩散动态。研究发现，具有更高“桥接资本”（bridging capital）的传播者更有利于后续采纳，而具有更高“结合资本”（bonding capital）的传播者则导致采纳减少；同时，俚语在更广泛语境中的使用与其被新用户正式采纳的时间呈正相关。这些结果表明，新词的传播不仅受社交网络结构的影响，也受到语言使用上下文的调节，揭示了社会结构与语言环境共同塑造在线新词扩散的核心机制。
链接: https://arxiv.org/abs/2609.25669
作者: Xiaoning Wang,Ted Underwood,Zhewei Sun
机构: University of Illinois Urbana-Champaign (伊利诺伊大学厄本那-香槟分校); Toyota Technological Institute at Chicago (芝加哥丰田技术学院)
类目: Computation and Language (cs.CL)
备注: Accepted to EMNLP 2026 main conference
Abstract:Adoption and diffusion of neologisms in online communities have received renewed attention in recent years. As internet slang terms such as APT, referring to a K-pop song, and phrases such as Canon Event meaning an embarrassing but pivotal event, go viral online, it becomes increasingly important to understand the mechanisms that contribute to their success. Prior studies have often explained slang diffusion either from the perspective of social interaction or from the linguistic properties of the slang itself, but rarely from both perspectives together. One major obstacle has been the high cost of annotating slang usage in large-scale online communication. Recent advances in large language models (LLMs), however, make it possible to use them as scalable annotators for such tasks. In this study, we first curate a human-annotated benchmark to evaluate LLM performance in detecting slang usage in real Reddit communication. We then leverage LLM-based annotations to model slang adoption and diffusion. Our results show that slang diffusers with higher bridging capital are associated with increased subsequent adoption, whereas diffusers with higher bonding capital are associated with reduced adoption. We also find that wider contextual usage of a slang term is associated with a longer time before new users officially adopt it. Together, these findings suggest that both social-network structure and linguistic context shape the diffusion of neologisms in online communities.
[NLP-57] Efficient Cost-Aware LLM Evaluation via Bayesian Bandit Gittins Indices ICML2026
【速读】： 该论文旨在解决大语言模型（Large Language Model, LLM）配置选择中因全面评估所有候选配置在每个基准测试项上所带来的高昂计算成本问题。现有方法在面对大规模候选配置或复杂基准时，难以在有限资源下高效识别高性能配置。为此，论文提出GittinsEval，将配置选择建模为一种成本感知的贝叶斯多臂老虎机（Bayesian bandit）问题，其核心在于采用贝叶斯最优的吉廷斯指数（Gittins index）策略，动态决定下一步评估的配置以及何时停止评估。该方案的关键创新在于引入一种任意时间（anytime）推荐规则，能够对已完全和部分评估的配置进行统一排序，并通过类似置信下界（LCB）的评分机制量化后验不确定性，从而在探索与利用之间实现高效平衡。GittinsEval具有计算效率优势，仅需离线预计算后的轻量级在线更新，即可在GSM8K、PIQA、AlpacaEval和MMLU等多类基准上持续表现出色，尤其在大规模样本基准上显著优于基于配置层面的贝叶斯优化，在大规模候选任务中亦超越无成本意识的多臂老虎机基线。关键成果表明，GittinsEval通常仅需1%至2%的全评估成本即可实现接近零的简单遗憾（simple regret），且自适应停止规则多在1%至10%的评估预算内触发，极大提升了配置搜索的效率与实用性。
链接: https://arxiv.org/abs/2609.25645
作者: Qian Xie,Yueli He,Nairen Cao
机构: Cornell University (康奈尔大学); Columbia University (哥伦比亚大学); New York University (纽约大学); Shanghai University of Finance and Economics (上海财经大学)
类目: Machine Learning (cs.LG); Computation and Language (cs.CL); Machine Learning (stat.ML)
备注: Spotlight at ICML 2026 Workshop on Decision-Making from Offline Datasets to Online Adaptation: Black-Box Optimization to Reinforcement Learning (DEMO)
Abstract:Exhaustively evaluating every candidate LLM configuration on every benchmark item to identify a high-performing one is costly. We formulate configuration selection as a cost-aware Bayesian bandit problem and propose GittinsEval, which draws on the Bayesian-optimal Gittins policy to determine which configuration to evaluate next and when to stop. We extend the policy with an anytime recommendation rule over both fully and partially evaluated configurations, using an LCB-style score to account for posterior uncertainty. GittinsEval is computationally efficient, requiring only lightweight online updates after offline precomputation. Across GSM8K, PIQA, AlpacaEval, and MMLU response matrices, GittinsEval is consistently competitive, with particularly strong gains over configuration-level Bayesian optimization on large-example benchmarks and over cost-unaware bandit baselines on large-candidate tasks. Crucially, GittinsEval often attains near-zero simple regret using only 1% to 2% of the exhaustive-evaluation cost; it also offers an adaptive stopping rule that typically triggers at 1% to 10%.
[NLP-58] Qwen 3.8-Omni: Towards Native Omni-Modal Agents
【速读】： 该论文旨在解决现有通用多模态模型在真实世界多模态生产力应用中面临的多模态理解与推理能力不足、长时序智能体任务执行能力薄弱，以及缺乏原生音频与视频支持的系统性挑战。其核心解决方案在于提出一种原生多模态联合训练策略（native multimodal co-training strategy），在保持强大文本领域能力的同时，有效实现从文本到音频与视频任务的智能体能力迁移。该模型基于Qwen3.8-Next的稀疏专家混合（MoE）架构，并将上下文窗口扩展至一百万令牌，显著增强长上下文多模态推理与长周期规划能力。为支撑实时多模态交互，研究进一步构建了系统级框架Qwen-Live-Harness，集成上下文与记忆管理、工具调用及子智能体委派等机制。配套开源工具Qwen-MM-Plugins则填补了现有智能体框架在原生音视频支持上的空白。实证评估表明，Qwen3.8-Omni-Flash在多模态理解、推理、长周期智能体执行及视频生产力任务上均表现卓越，为科研与生产环境中部署原生多模态智能体提供了可落地的技术基础。
链接: https://arxiv.org/abs/2609.25611
作者: Qwen Team
机构: Qwen Team
类目: Computation and Language (cs.CL); Computer Vision and Pattern Recognition (cs.CV); Multimedia (cs.MM)
备注:
Abstract:We introduce Qwen3.8-Omni-Flash, a natively multimodal agentic model for real-world multimodal productivity. Compared with previous omni models, which primarily emphasized perception and interaction, Qwen3.8-Omni-Flash substantially improves multimodal understanding and reasoning, as well as performance on long-horizon agentic tasks. These capabilities are supported by a native multimodal co-training strategy that preserves strong text-domain capabilities while facilitating the transfer of agentic capabilities from text to audio and video tasks. The model inherits the sparse mixture-of-experts (MoE) architecture of Qwen3.8-Next and extends the context window to one million tokens, supporting long-context multimodal reasoning and long-horizon planning. These advances enable integration into production workflows as a primary agent or a specialized sub-agent, supporting video editing, long-form audio and video translation, music-conditioned music video or movie generation, and video-based note or omni-skill creation. To address the lack of native audio and video support in existing agent harnesses, we release Qwen-MM-Plugins, a lightweight open-source plugin framework for multimodal productivity. We further frame real-time multimodal interaction as a system-level challenge requiring orchestration of context and memory management, tool use, and sub-agent delegation. Accordingly, we release Qwen-Live-Harness, an open-source framework for building responsive, real-time multimodal agents based on Qwen3.8-Omni-Flash. Extensive evaluations demonstrate that Qwen3.8-Omni-Flash achieves strong performance across multimodal understanding, reasoning, long-horizon agentic execution, and video productivity tasks. These results and the accompanying open-source tools support Qwen3.8-Omni-Flash as a practical foundation for deploying natively multimodal agents in research and production.
[NLP-59] Rewired or Gated? How Instruction Tuning Shapes Knowledge-Conflict Circuits in LLM s EMNLP2026
【速读】： 该论文旨在解决指令微调（instruction tuning）如何影响语言模型在提示（prompt）与模型权重之间冲突时的决策机制这一核心问题。具体而言，研究关注的是：在出现矛盾信息时，模型是通过重构其内部的冲突解决电路，还是仅对已有电路进行门控或权重调整。研究采用五种独立的可解释性方法（节点与边归因、超位置角色分析、因果消融、路径修补等），对三种主流模型家族（Llama-3.2-3B、Qwen-2.5-3B、Gemma-3-4B）进行了基线模型与指令微调模型的对比分析，发现冲突解决的关键机制并非被替换，而是由相同的后层注意力头（late-layer attention heads）以重加权的方式实现动态调控，且节点重叠度高达0.60–0.82，表明其本质为“门控”而非“重构”。行为层面，指令微调使模型更倾向于依赖参数记忆（parametric memory），对简短的反事实上下文表现出更强的排斥性，这与用户应遵循提示的直觉相反；但这种增强的怀疑性具有显著的框架依赖性——当虚假陈述以连贯、具证据性的段落形式呈现时，该效应即消失。因此，指令微调带来的鲁棒性虽真实存在，但仅限于特定攻击模式。更广泛地，由于冲突解决电路在微调前后保持结构一致性，研究认为针对基线模型开发的可解释性与控制工具可直接迁移至指令微调后的部署版本。
链接: https://arxiv.org/abs/2609.25602
作者: Shubham Santosh Pandere,Gautam Ranka,Ritika Varshney,Navya Deshmukh,Roushni Sareen,Roshan Kumar Singh
机构: IvLabs, VNIT(IV实验室，文达纳理工学院); Google(谷歌)
类目: Machine Learning (cs.LG); Computation and Language (cs.CL)
备注: Accepted at BlackboxNLP 2026, Co-located with EMNLP 2026
Abstract:In language models, the choice between believing the prompt and believing the weights is made by a handful of identifiable attention heads. Instruction tuning changes how models behave under conflict, but whether it rewires the underlying circuit or merely gates/reweights already present components, remains unknown. We provide the first mechanistic base-vs-instruct comparison of conflict-resolution circuits, across three families (Llama-3.2-3B, Qwen-2.5-3B, Gemma-3-4B). Five independent methods, node and edge attribution, superposition role analysis, causal ablation, and path patching, converge on gating, with the same heads, in the same late-layers, are found to be reweighted rather than replaced with a high node overlap (0.60-0.82). Behaviorally, tuning shifts models toward parametric memory, making instruct models reject a terse counterfactual context far more than base ones, the opposite of a naive user-following expectation. Yet this added skepticism is a factor of framing since it disappears when the same false claim is delivered as a coherent, evidential passage. The robustness that instruction tuning buys against terse injection is therefore real but narrow. More broadly, we believe that because the conflict circuit is preserved rather than rebuilt, interpretability and control tools calibrated on base models should transfer directly to their deployed instruct siblings.
[NLP-60] Compressing Long Context into Answer-Aligned Memory Embeddings for LLM Inference
【速读】： 该论文旨在解决大语言模型（Large Language Model, LLM）在长上下文推理时面临的计算与内存瓶颈问题，具体表现为自注意力机制的二次方复杂度增长以及键值缓存（KV cache）的线性扩展导致的延迟上升、能耗增加和GPU显存占用激增。现有软压缩（soft-compression）方法普遍存在推理阶段缺乏查询引导的记忆选择、训练过程缺少以答案为目标的监督信号，或压缩策略与特定解码器架构过度耦合等缺陷。为此，本文提出一种上下文到答案对齐的记忆压缩框架（Context-to-Answer-Aligned Memory Compression, CMC），其核心在于将长输入上下文压缩为与任意冻结解码器嵌入空间对齐的紧凑上下文记忆嵌入（Context Memory Embeddings, CMEs），从而在不修改解码器权重的前提下降低推理开销。CMC的关键创新包括：引入双层KV缓存结构，融合基于问题引导的CME选择与局部上下文窗口；采用从冻结的LLM进行答案目标导向的蒸馏训练压缩器。实验在九种编码器-解码器组合及四个问答基准上验证了CMC的有效性，结果显示其在SQuAD上最高实现7.3的精确匹配（EM）与4.0的F1提升，同时在生成3000个标记时将推理时间与能耗降低最多20%，峰值显存占用减少达50%。消融实验进一步证实了各组件与训练目标对性能的贡献。
链接: https://arxiv.org/abs/2609.25537
作者: Md Mostafizer Rahman,Md Faizul Ibne Amin,Md Shahajada Mia,Yutaka Watanobe,Fang Liu
机构: Lucy Family Institute for Data Society, University of Notre Dame, IN, USA; The University of Aizu, Aizuwakamatsu, Japan; Department of Applied and Computational Mathematics and Statistics, University of Notre Dame, IN, USA
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
备注:
Abstract:Large language model (LLM) inference is constrained by the quadratic scaling of self-attention and the linear scaling of the KV cache, increasing latency, energy consumption, and GPU memory demand as context length scales. Existing soft-compression methods either lack query-guided memory selection at inference time, train without answer-targeted supervision, or couple compression tightly to a specific decoder architecture. We propose a Context-to-Answer-Aligned Memory Compression (CMC) framework, which compresses long input contexts into compact Context Memory Embeddings (CMEs) aligned to any frozen decoder’s embedding space, reducing inference costs without modifying decoder weights. CMC introduces a two-tier KV cache that combines question-guided CME selection with a local context window, and trains the compressor with answer-targeted distillation from a frozen LLM. Experiments across nine encoder-decoder combinations and four QA benchmarks show that CMC consistently outperforms the baseline, achieving up to 7.3 EM and 4.0 F1 point gains on SQuAD, while reducing inference time and energy consumption by up to 20% and peak reserved GPU memory by up to 50% at 3,000 generation tokens. Ablation studies confirm that each architectural component and training objective contributes to the performance.
[NLP-61] Matryoshka attribution: Learning to attribute language model outputs to representations and weights
【速读】： 该论文旨在解决大语言模型（Large Language Models, LLMs）输出可解释性中的核心问题——如何准确归因于模型内部计算过程。现有方法如基于因果干预、梯度或可学习掩码的策略，或计算成本过高，或难以识别真正具有因果重要性的内部组件。本文提出将归因问题重新定义为寻找能够最小化下游损失的嵌套内部组件子集，从而实现更精准的因果定位。其解决方案的关键在于提出一种名为“套娃归因”（Matryoshka Attribution, MAttr）的掩码学习方法，该方法采用可微分的sigmoid top-k算子对掩码进行参数化，并通过在训练过程中随机化top-k阈值（k）来同时监督所有稀疏度下的模型训练，从而学习到组件按归因得分排序的结构化顺序。MAttr在机制可解释性基准测试（Mechanistic Interpretability Benchmark）中取得官方排行榜第一名，能够识别出稀疏且具备任务迁移能力的神经电路。此外，该方法还可结合强化学习应用于微调过程中的权重变化分析：实验表明，在Llama 3.1 8B Instruct模型上仅恢复1%的权重至基线模型状态，即可有效消除拒绝行为，同时保持原有能力。因此，MAttr成功将可解释性问题转化为一个可通过梯度下降优化的可学习目标，为未来研究提供了新的范式。
链接: https://arxiv.org/abs/2609.25518
作者: Aryaman Arora,Kirill Acharya,Nathan Hu,Yanzhe Zhang,Noah Goodman,Dan Jurafsky,Christopher Potts
机构: Stanford University(斯坦福大学)
类目: Computation and Language (cs.CL); Machine Learning (cs.LG)
备注: 10 pages main text, 58 pages total; preprint
Abstract:Attributing language model outputs to their internal computations is an open problem in interpretability. Existing methods, which use causal interventions, gradients, or learnable masks, either are infeasibly expensive or struggle to identify actual causally-important internal computations. We propose framing attribution as the problem of identifying nested subsets of internal components which minimise a downstream loss. To learn this task, we introduce Matryoshka Attribution (MAttr), a mask learning method that parametrises the mask with a simple differentiable sigmoid top- k operator. We supervise training over all sparsities simultaneously by randomising k over training, resulting in a learned ordering of components by attribution score. MAttr achieves number 1 on the official leaderboard of the Mechanistic Interpretability Benchmark (Mueller et al., 2025); our method identifies sparse and task-transferrable circuits across varying circuit bases. As a practical application, we show that MAttr can be trained with reinforcement learning to identify weight changes responsible for downstream behaviours in LLM finetuning. We train MAttr on refusal judge scores and find that restoring 1% of Llama 3.1 8B Instruct’s weights to their base model state is sufficient to remove refusals while maintaining capabilities. We view MAttr as a successful formulation of interpretability into a learnable objective that we can tackle with gradient descent, and encourage future work along these lines.
[NLP-62] Universal Fractal Natural Language Decision Map: Real-Time Edge Triage Across Heterogeneous Domains
【速读】： 该论文旨在解决大语言模型（Large Language Models, LLMs）在运行时操作诊断（runtime operational triage）中面临的高延迟（100–500 ms）、高显存占用（4–8 GB VRAM）及能耗过大的核心瓶颈问题。其解决方案的关键在于提出一种基于曼德博集合混沌边界动态调制的通用分形自然语言决策图（Universal Fractal Natural Language Decision Map），通过无权重张量（0 Bytes VRAM）的werr机器原生边缘反射式运行时与生产级answer平台实现，利用24字节坐标种子在曼德博集合的混沌边界上动态调节并评估四象限逃逸动力学，生成确定性决策（布尔型noul、分类型choice、序数型score）。该方法受生物系统一型反射弧启发，引入三项核心技术：（i）基于领域投影器Phi_D的自动种子路由机制，相较线性基线提升28.8%准确率；（ii）基于信息论的声学阻尼滤波器，结合词元熵与语音频谱密度，有效抵御提示注入攻击（实测绕过率为0.0%，95%置信区间[0.0%, 30.8%]），同时减少逃逸迭代45.8%，将吞吐量加速至2.5倍，延迟降至3.31 ms；（iii）采用O(1)指数移动平均（EMA, alpha=0.03）与象限相位旋转的有机动态校准框架，消除位置偏差。在裸金属硬件上对超过1,150项验证决策（3,200+问题）进行基准测试，于独立JevBench评测中位列全球第一（81.65%得分），实现92.6%宏平均准确率（95%置信区间[90.8%, 94.1%]），7.08 ms中位数CPU延迟，并提供兼容OpenAI API接口（/v1/chat/completions），成功验证其在微控制器及32字节EVM智能合约上的可行性。
链接: https://arxiv.org/abs/2609.25498
作者: Volkan Dağlı,Zerrin Dağlı,Dağhan Dağlı
机构: Anadolu University (安纳托利亚大学); ITouch Systems (伊托奇系统); Mersin University (梅尔辛大学); Toros Science College (托罗斯科学学院)
类目: Neural and Evolutionary Computing (cs.NE); Artificial Intelligence (cs.AI); Computation and Language (cs.CL)
备注: 10 pages, 5 figures, 3 tables. Companion to Mandelbrot Fractal Neural Synthesis. Live portal: this https URL ; Source code: this https URL
Abstract:Deploying Large Language Models for runtime operational triage incurs prohibitive latency (100-500 ms), high VRAM requirements (4-8 GB), and excessive energy dissipation. Extending Mandelbrot Fractal Neural Synthesis (Dagli et al., 2026), this paper presents the Universal Fractal Natural Language Decision Map, realized via the werr machine-native edge reflex runtime and the production answerr platform (this https URL). Operating entirely without stored weight tensors (0 Bytes VRAM), the engine synthesizes deterministic decisions—noul (Boolean), choice (categorical), and score (ordinal)—by dynamically modulating 24-byte coordinate seeds along the chaotic boundary of the Mandelbrot set and evaluating 4-quadrant escape dynamics. Drawing inspiration from biological System-One reflex arcs, the engine introduces: (i) an Auto-Seed Router with domain projector Phi_D yielding a +28.8% accuracy gain over linear baselines; (ii) an Information-Theoretic Acoustic Damping Filter grounded in token entropy and phonetic spectral density that insulates against prompt injections (0.0% empirical bypass; 95% Wilson CI: [0.0%, 30.8%]) while pruning escape iterations by 45.8% (accelerating throughput 2.5x to 3.31 ms latency); and (iii) an Organic Dynamic Calibration framework using O(1) Exponential Moving Average (EMA, alpha=0.03) and quadrant phase rotation to eliminate positional bias. Benchmarked on bare-metal infrastructure (this http URL) across 1,150+ verified decisions (3,200+ questions) and ranked World #1 on the independent JevBench suite (81.65%), the framework achieves 92.6% macro-accuracy (95% CI: [90.8%, 94.1%]) with 7.08 ms median CPU latency. We provide an OpenAI-compatible API (/v1/chat/completions) and demonstrate feasibility on microcontrollers and 32-byte EVM smart contracts.
[NLP-63] Mining Legal Arguments in U.S. Corporate Case Law
【速读】： 该论文旨在解决美国联邦税法判例中结构化法律论证挖掘的难题，特别是针对《国内税收法典》第368条项下公司重组案件的论证结构识别与建模问题。其核心挑战在于如何准确标注并建模法律文本中论证单元（如规则、分析、结论等）之间的功能关系及支持路径，从而实现对法律推理过程的系统性解析。解决方案的关键在于构建首个由专家标注的树状结构论证语料库（tree-structured argument corpus），包含42份美国联邦税法判决书，明确标注五类功能单元：规则（Rule）、分析（Analysis）、结论（Conclusion）、背景事实（Background Facts）和程序历史（Procedural History）。其中，规则、分析与结论可构成有向支持树，而背景事实与程序历史则提供上下文支撑。该数据集同时提供基于片段、句子、扁平化及树形结构的多层级表示形式，为法律段落分类、论证检索与推理补全任务提供基准支持。实证研究表明，功能标签在跨案例评估中具备可学习性，监督微调能提升同案内检索性能，但跨案例泛化能力仍较弱，反映出当前模型在捕捉深层论证结构上的局限性。该语料库为美国联邦税法判例中的结构化论证挖掘提供了保守且可靠的基准。
链接: https://arxiv.org/abs/2609.25441
作者: Luis Brena,William Jurayj,Gregory Deyesu,Zaid Al-Huneidi,Andrew Blair-Stanek,Benjamin Van Durme
机构: Johns Hopkins University (约翰霍普金斯大学); University of Maryland School of Law (马里兰大学法学院)
类目: Computation and Language (cs.CL)
备注: 28 pages, 4 figures
Abstract:Legal argument mining supports passage classification, retrieval, and argument completion. This work introduces an expert-annotated dataset of 42 U.S. federal tax opinions on corporate reorganizations under I.R.C. §368. To our knowledge, it is the first expert-annotated, tree-structured argument corpus for this domain. Explicit spans receive one of five functional labels: Rule, Analysis, Conclusion, Background Facts, and Procedural History. Rule, Analysis, and Conclusion spans can be linked into directed support trees, while Background Facts and Procedural History serve a contextual function. The corpus provides span-based, sentence-based, flat, and tree-structured representations. Agreement analysis shows that functional node labels are more reliable than directed support edges and implicit intermediate conclusions. Directed-path agreement is stronger than direct-edge agreement, which indicates that broad reachability is more stable than exact local decomposition. Classification experiments show that functional labels are learnable under case-disjoint evaluation. Retrieval experiments show that supervised fine-tuning improves within-case retrieval. However, cross-case generalization remains weak. The dataset supports legal passage classification and provides a conservative benchmark for structured argument mining in U.S. federal tax case law.
[NLP-64] Efficient Iterative Retrieval with Heterogeneous Batching EMNLP2026
【速读】： 该论文旨在解决现代信息检索系统中嵌入模型（Embedding Model）与生成式 AI（Generative AI）在推理服务阶段因孤立执行导致的吞吐量低、GPU 利用率差的问题。现有系统通常采用粗粒度划分（如将 GPU 固定分配给特定任务），难以适应动态负载变化，进而产生计算“空洞”（computational bubbles）。其核心解决方案在于提出 Orthrus，一个在统一推理循环内实现异构任务混合批处理（Heterogeneous Batching）的服务系统。关键创新包括：基于分块嵌入（Chunked Embedding）与增量池化（Incremental Pooling）机制，有效协调嵌入与生成任务间冲突的计算模式；同时通过工作负载感知的批量组成调整策略，动态优化批处理结构以提升性能。实验结果表明，在四块 A100 GPU 上，相较于基线部署，Orthrus 在受控工作负载下实现了 1.28×–4.52× 的吞吐量提升，并在迭代式检索增强生成（iterative-RAG）基准测试中将端到端 p99 延迟降低高达 55.8%。
链接: https://arxiv.org/abs/2609.25405
作者: Dohyun Park,Hubertus Franke,Daniel G. Waddington,Swaminathan Sundararaman,Yongjoo Park
机构: 未知
类目: Artificial Intelligence (cs.AI); Computation and Language (cs.CL)
备注: 15 pages, 8 figures, Accepted to EMNLP 2026 (main conference)
Abstract:Modern information retrieval increasingly employs both embedding and generative models to handle complex queries. However, current serving systems suffer from low throughput and poor GPU utilization because they execute these models in isolation. Coarse-grained partitioning, such as dedicating GPUs to specific tasks, fails to adapt to dynamic workloads and creates computational “bubbles”. To address these, we present Orthrus, a serving system that performs heterogeneous batching within a unified inference loop. The primary challenge lies in unifying embedding and generation workloads with conflicting computational patterns while optimizing batch composition for high performance. Orthrus addresses these challenges through chunked embedding with incremental pooling and by adjusting batch composition in a workload-aware manner. Evaluation on four A100 GPUs shows that, relative to baseline deployments, Orthrus achieves 1.28 \times --4.52 \times higher throughput on controlled workloads and up to 55.8% lower end-to-end p99 latency on an iterative-RAG benchmark. We release our code at this https URL .
[NLP-65] Passes Alone Fails Together: Benchmarking Semantic Coordination in Parallel LLM -Agent Development
【速读】： 该论文旨在解决并行生成式代码代理（parallel coding agents）在独立生成代码片段后合并时出现语义协调失败的问题，其核心挑战在于：当一个代理修改了另一个代理仍依赖的接口或规则时，虽各片段单独运行正常，但合并后产生错误。解决方案的关键在于引入“stale”这一基准测试框架，用于系统评估代码片段在合并过程中引入的语义不一致问题。研究通过三类实验层级验证：受控接口变更的合成任务、真实GitHub Pull Request对的合并测试，以及基于12个真实Django工具函数构建的复杂任务。结果显示，在417组经挖掘的Django代码对中，仅1例在修正评分流程后显示合并干扰；而在构造的任务中，干扰率高达97%。进一步发现，通过向并发修改提供明确的完成状态消息，可恢复82%的失败场景。研究揭示，尽管审查中的代码合并请求看似未包含大量并行冲突，但实际在真实代码环境中，此类语义协调问题仍普遍存在，且现有失败率无法准确反映其在实践中的发生频率。
链接: https://arxiv.org/abs/2609.25396
作者: Haocheng Xia,Eugene Wu,Yongjoo Park
机构: University of Illinois Urbana-Champaign (伊利诺伊大学厄本那-香槟分校); Columbia University (哥伦比亚大学)
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI); Software Engineering (cs.SE)
备注: 6 pages, accepted to The 2nd Workshop on Explainable and Reliable Software Systems (EXPRESS 2026)
Abstract:Parallel coding agents can produce patches that work alone but fail when merged. This happens when one agent changes an interface or rule that another agent still relies on. We study these failures with stale, a benchmark for semantic coordination. Our evaluation runs the same tests on each patch alone and on their combination, counting only failures introduced by combining the patches. We use three tiers: synthetic tasks with controlled interface changes, pairs of merged pull requests, and constructed tasks that use real Django helpers. Among 834 runs on 417 mined Django pairs, only one showed interference after correcting the grading procedure. On constructed tasks using 12 Django helpers, interference occurred in 97% of runs. A message describing the completed concurrent change recovered 82% of runs. Reviewed pull requests may contain few unresolved parallel changes, even when agents fail on controlled tasks using real code. The constructed failure rates do not estimate how often these problems occur in practice.
[NLP-66] comGPT -R1: Unified Post-Training for Reasoning Across Heterogeneous Telecom Tasks
【速读】： 该论文旨在解决现有电信领域大语言模型（LLM）在跨多类型任务与数据源时推理能力不足的问题。具体而言，通用大模型缺乏对电信专业领域的可靠知识锚定，而专用模型则通常局限于狭窄的任务范畴，难以实现多任务泛化。为应对这一挑战，论文提出TelecomGPT-R1，一个基于协议（protocol）、知识（knowledge）、建模（modeling）和故障（fault）四大互补维度构建的开源统一电信推理模型家族。其核心解决方案在于：首先设计一种轴向感知（axis-aware）的数据生成框架，将粗粒度公开电信资料转化为经验证的问答对及高质量思维链（Chain-of-Thought, CoT）推理轨迹，形成包含104,880个样本的训练语料库；随后通过监督微调（SFT）注入电信知识与证据驱动的推理模式，突破强化学习（RL）的冷启动瓶颈；最后采用动态采样策略优化（DAPO）结合任务路由评分机制，在异构电信推理任务中保持强化学习更新的信息量与稳定性。该评分机制将轴向特定的CoT轨迹分解为可验证的推理单元，并融合基于证据的密集过程奖励与结果正确性，使模型能够从可验证的电信证据中学习可泛化的求解行为。实验表明，开源的TelecomGPT-R1-27B在GSMA开放电信基准测试集上取得89.64%的平均得分，超越包括GPT-5、Claude和Gemini在内的主流专有模型。
链接: https://arxiv.org/abs/2609.25356
作者: Bohao Wang,Chenwei Wu,Hang Zou,Yu Tian,Lina Bariah,Li Wei,Chongwen Huang,Yongliang Shen,Zhaoyang Zhang,Merouane Debbah
机构: Zhejiang University (浙江大学); University of Michigan (密歇根大学); Khalifa University (哈利法大学)
类目: Computation and Language (cs.CL)
备注:
Abstract:Large language models (LLMs) offer great potential to automate a broad range of telecom engineering tasks by reasoning over standards, network configurations, mathematical models, source code, and operational logs. However, existing telecom LLMs struggle to reliably reason across these diverse tasks and data types. General-purpose LLMs often lack reliable grounding in telecom-specific knowledge, while telecom-specialized models are typically developed for narrower task families and exhibit limited multi-task performance. To fill this gap, we introduce TelecomGPT-R1, a family of open source unified telecom reasoning models structured around four complementary axes: protocol, knowledge, modeling, and fault. We first develop an axis-aware data generation framework that refines coarse public telecom artifacts into verified question-answer pairs and high quality chain-of-thought (CoT) reasoning trajectories, yielding a training corpus containing 104,880 examples. Building on this corpus, supervised fine-tuning (SFT) instills telecom knowledge and evidence-grounded reasoning patterns to overcome the cold start barrier for reinforcement learning (RL). We then apply dynamic sampling policy optimization (DAPO) with task-routed rubric rewards to keep RL updates informative and stable across heterogeneous telecom reasoning tasks. These rewards decompose axis-specific CoT traces into verifiable reasoning units and combine grounded dense process credit with outcome correctness, allowing RL to learn generalizable problem solving behaviors from verifiable telecom evidence. We release the TelecomGPT-R1 models and a reproducible training recipe to support further community development. Evaluations on seven benchmarks of the GSMA Open Telco Leaderboard show that the open-source TelecomGPT-R1-27B achieves an 89.64% mean score, outperforming leading proprietary models, including GPT-5, Claude, and Gemini.
[NLP-67] FineWeb-CLaR: Culture Language and Region Annotations for Benchmark-Aligned Corpus Auditing EMNLP2026
【速读】： 该论文旨在解决语言模型在文化评估覆盖范围与鲁棒性方面难以诊断的问题，其核心挑战在于预训练语料库与文化基准数据集普遍缺乏可比的元数据标注。随着文化基准日益聚焦于语言、区域及地方性实践等文化情境化现象，而大规模网络语料库通常仅按语言组织，导致二者在文化维度上难以对齐。为此，论文提出关键解决方案——构建一个共享的文化-语言-区域层级（culture-language-region layer），实现不同资源间的可比性。其核心创新为引入FineWeb-CLaR，这是一个基于FineWeb与FineWeb-2的大规模标注数据集，通过URL推断的区域标签和文化主题溯源，将309亿条网页文档映射至统一的文化-语言-区域坐标轴上，支持语料库审计与基准对齐。该数据集为25.61%的文档（79.2亿条）分配了非空区域标签，并基于刘等（2025）提出的文化分类体系，通过推导地方性主题并投影至14个分类叶节点，生成可用于语料侧比较的“地方主题分布”（Locale Topic Distributions, LTDs）。同时，论文还对277个文化自然语言处理基准进行同源分类学标注，涵盖相同语言与区域覆盖范围。这一系统性框架实现了语料预训练证据与基准评估覆盖之间的直接对比，显著提升了文化评估的透明度与可审计性。
链接: https://arxiv.org/abs/2609.25298
作者: Yusser Al Ghussin,Eva Gavaller,Cristina España-Bonet,Josef van Genabith,Simon Ostermann
机构: Saarland University (萨尔兰大学); German Research Center for Artificial Intelligence (DFKI)(德国人工智能研究中心); Barcelona Supercomputing Center (BSC-CNS)(巴塞罗那超级计算中心)
类目: Computation and Language (cs.CL)
备注: accepted to EMNLP 2026 (Main)
Abstract:Cultural evaluation coverage and robustness in language models are difficult to diagnose because pretraining corpora and cultural benchmarks are rarely indexed with comparable metadata. Benchmarks increasingly target culturally situated phenomena at the level of languages, regions, and locale-specific practices, while web-scale corpora are usually organized only by language. A shared culture-language-region layer makes these resources comparable, enabling audits of whether a target cultural phenomenon is represented in pretraining data, evaluated by benchmarks or both. To this end, we introduce FineWeb-CLaR, a large-scale annotated dataset derived from FineWeb and FineWeb-2 that places web documents on a shared culture-language-region axis for corpus auditing and benchmark alignment. FineWeb-CLaR annotates the full 30.9B-document collection from FineWeb and FineWeb-2 with URL-derived region labels and cultural-topic provenance. Our region resolver assigns a non-empty region to 25.61% of documents (7.92B). For cultural-topic analysis, we induce locale-specific topics and project them onto the 14 leaves of the Cultural Taxonomy of Liu et al. (2025), producing Locale Topic Distributions (LTDs) for corpus-side comparison. We also annotate 277 cultural NLP benchmarks with the same taxonomy, language coverage, and region coverage. Together, these resources enable direct comparison between corpus-side pretraining evidence and benchmark-side evaluation coverage. Comments: accepted to EMNLP 2026 (Main) Subjects: Computation and Language (cs.CL) Cite as: arXiv:2609.25298 [cs.CL] (or arXiv:2609.25298v1 [cs.CL] for this version) https://doi.org/10.48550/arXiv.2609.25298 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
[NLP-68] rains but Doesnt Learn: A Post-Training Delivery Benchmark for LLM Agents as Forward-Deployed Engineers EMNLP2026
【速读】： 该论文旨在解决生成式AI在后训练服务化（Post-training as a Service, PTaaS）场景下的可信交付问题，即如何确保由大语言模型（LLM）代理担任远程部署工程师（Forward-Deployed Engineer, FDE）时，其生成的微调模型不仅在指标上表现提升，更能在实际运行中真正实现有效学习并满足可复现性、预算约束及人工审批等治理要求。其核心挑战在于“训练但未学习”（Train but Do Not Learn, TBDL）这一隐蔽失败模式：尽管损失函数下降、各项监控信号保持正常，最终模型性能却与基础模型无异。解决方案的关键在于构建一个受控的治理交付平面（governed delivery plane），通过引入基于平台记录事实的全程阶段评分机制（oracle scoring across ten stages）和由人工FDE作为参照基准的对比评估体系，实现对每个训练流程的透明追踪与质量验证。此外，研究设计了一种基于已知污染样本校准的运行中检测器，在训练过程中实时识别严重数据或配置污染，从而在支付前通过操作员主导的验收门禁机制拦截所有失败案例，显著提升了自动化微调流程的可靠性与可信度。
链接: https://arxiv.org/abs/2609.25237
作者: Weihang Ding,Junfei Zhan
机构: University of California, Berkeley (加州大学伯克利分校); Imperial College London (帝国理工学院)
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Computation and Language (cs.CL)
备注: 12 pages, 3 figures. Accepted to EMNLP 2026 Industry Track
Abstract:Post-training is becoming a service (PTaaS): a customer hands an operator data and a goal, and a forward-deployed engineer (FDE) returns a fine-tuned, evaluated, and deployed model under a budget, a human-approval gate, and reproducibility requirements. Seating an LLM agent in the FDE seat raises a question existing benchmarks cannot answer: not whether an agent can raise a metric, but whether it can be trusted to deliver. We answer it on a governed delivery plane, where an agent drives ten stages and an oracle scores each stage from platform-recorded facts. The central silent failure is the run that trains but does not learn (TBDL): loss falls, every signal stays green, and the delivered model is no better than the base. An operator-run acceptance gate catches every such run before payment, and a detector calibrated on known-corrupted runs flags severe corruption mid-run. We ran four frontier agents (Claude Opus 5, GPT-5.6-luna, Gemini 3.7 Flash, DeepSeek V4-Pro) end to end on metered L40S, A100, and H200 GPUs across 8B to 70B open bases, certifying every scenario before scoring. We also ran a human FDE arm under the same oracle and compare every agent against it.
[NLP-69] FinFIRST: Benchmarking Search Agents for Financial Information Retrieval Sourcing and Traceability
【速读】： 该论文旨在解决大语言模型（LLM）在金融领域任务中普遍存在“黑箱式”输出的问题，即现有评估基准仅关注最终答案的正确性，而忽视了回答背后的证据链完整性、信息时效性、数据源权威性、实体与时间周期对齐、单位与定义一致性以及结论可验证性等关键环节。其核心挑战在于缺乏一个能够同时衡量答案质量与支撑证据可信度的系统性评估框架。为此，研究提出FinFIRST（Financial Information Retrieval, Sourcing and Traceability），作为首个联合评估答案与支持证据的金融领域基准。其解决方案的关键在于构建一套基于原子级评分标准（atomic rubrics）的多维度评估体系，涵盖原始信息获取、来源验证、计算与答案生成三个维度，并通过123个由50余位金融专家共同设计的任务，覆盖18个金融领域分类和六轴覆盖蓝图，结合138个权威金融数据源及六阶段质量控制流程，实现对模型推理过程的可测量、可验证与可诊断。实验表明，尽管部分模型在最终答案正确率上表现良好，但在计算与答案形成环节仍显著落后于原始信息获取能力，凸显了当前模型在复杂金融推理中对证据整合与逻辑推演的不足。
链接: https://arxiv.org/abs/2609.25192
作者: Wenqing Wang,Haitao Xiang,Xinyi Zhao,Mingming Yin,Ying Zhong,Zhaoxin Huan,Qiheng Zhou,Jin Zhu,Xiaolu Zhang,Shi Chang,Jun Zhou
机构: Ant Group (蚂蚁集团)
类目: Computation and Language (cs.CL)
备注: 20 pages, 3 figures, and 7 tables. Dataset available at this https URL
Abstract:Financial search is a highly demanding task for LLM agents, requiring not only a correct final answer but also temporally valid information retrieval, authoritative source selection, entity and period alignment, unit and definition consistency, and verifiable evidence for all conclusions. Existing benchmarks predominantly evaluate only the final answer, making it difficult to localize errors or assess whether an answer is well-founded. To address this gap, we introduce FinFIRST (Financial Information Retrieval, Sourcing and Traceability), the first financial benchmark to jointly evaluate answers and supporting evidence through atomic rubrics. FinFIRST comprises 123 expert-authored tasks spanning a graduated difficulty spectrum, constructed from aggregate patterns of real-world financial scenarios through an 18-field taxonomy, a six-axis coverage blueprint, a registry of 138 financial sources, contributions from over 50 finance experts, and a six-stage quality-control pipeline. Each task is accompanied by an evidence-grounded reference package decomposed into atomic criteria across three dimensions: raw-information acquisition, source verification, and computation and answer formation. We evaluate 15 model configurations under a unified tool setting. Claude-Opus-5 achieves the highest atomic score of 87.59%, while GPT-5.6-Sol attains the highest strict pass rate of 71.54%. Computation and answer formation consistently lag behind raw-information acquisition across systems. FinFIRST retains final-answer correctness as the primary objective while making the supporting research process measurable, verifiable, and diagnosable.
[NLP-70] From Pattern Recognizers to Personalized Companions: A Survey of Large Language Models in Mental Health
【速读】： 该论文旨在解决生成式 AI 在心理健康支持领域应用中缺乏系统性演进框架的问题，即当前研究虽迅速增长但呈现碎片化特征，难以厘清技术发展脉络与未来方向。其核心解决方案的关键在于提出一个三阶段演化模型：第一阶段（Phase I）中，大语言模型（LLM）作为被动的信息工具与模式识别器用于心理评估；第二阶段（Phase II）中，模型演变为具有共情能力的对话代理，支持即时、无状态的交互；第三阶段（Phase III）则聚焦于构建具备状态记忆的长期个性化陪伴者，即基于认知代理架构（包含身份表征、记忆、推理与规划模块）的持续性智能体。该框架通过系统梳理关键技术、代理架构及数据集与评测基准的演进，揭示了从功能辅助到深度陪伴的技术跃迁路径，为实现负责任、高效且以人为本的心理健康AI提供了清晰的发展蓝图。
链接: https://arxiv.org/abs/2609.25186
作者: He Hu,Yucheng Zhou,Qianning Wang,Yingjian Zou,Chiyuan Ma,Juzheng Si,Jianzhuang Liu,Zitong Yu,Laizhong Cui,Fei Ma,Qi Tian
机构: 未知
类目: Computers and Society (cs.CY); Artificial Intelligence (cs.AI); Computation and Language (cs.CL)
备注:
Abstract:The rising global prevalence of mental health conditions, together with longstanding barriers in traditional healthcare, such as limited resources, high cost, stigma, and privacy concerns, has created an urgent need for accessible and scalable support. Large Language Models (LLMs) have emerged as a transformative technology with strong potential to democratize mental health support through advanced natural language understanding and generation. However, the rapidly expanding, fragmented body of work in this area lacks a coherent evolutionary narrative, making it difficult to contextualize current progress and identify future directions. This survey addresses this gap by organizing and analyzing the literature around a central thesis: the role of LLMs in mental health is evolving through three distinct, increasingly sophisticated phases. We trace this trajectory from Phase I, in which LLMs act primarily as passive Information Tools and Pattern Recognizers for assessment; through Phase II, where they function as Empathetic Conversationalists for in-the-moment, stateless interactions; to the current frontier, Phase III, which seeks Longitudinal, Personalized Companions implemented as stateful cognitive agents. To support this framework, we systematically review core technologies, agent architectures (Profile, Memory, Reasoning, and Planning), and the critical infrastructure of datasets and benchmarks, highlighting how their evolution underpins this developmental path. Viewing the field through this developmental lens, we provide a comprehensive synthesis of existing work, an insightful narrative of its trajectory, and a clear roadmap for future innovation in responsible, effective, and human-centered AI for mental healthcare. A curated collection of the resources reviewed in this survey is available at our project repository: this https URL.
[NLP-71] Impact Is Not Invalidation: Ask About the Claim Not the Diff
【速读】： 该论文旨在解决编码代理（coding agents）在存储的断言（claim）因代码库变更而可能失效时，如何准确判断哪些断言已不再成立的问题。现有方法中，基于内容锚定（content anchoring）的机制会因源代码任意变动即触发失效判断，导致误报频发；而依赖语义等价性分类（semantic-equivalence classification）来判断变更是否保持行为不变，其有效性受制于问题表述方式而非模型本身能力。研究发现，当询问“某个提交是否保持行为不变”时，五种跨越40倍价格范围的模型在真实提交上仍触发率达59%-72%，且精确率仅为0.291至0.329（基线为0.25），表现不佳；而将问题改为“某一特定断言是否仍然有效”，相同模型在同一差异上精确率显著提升至0.705至0.974。进一步实验表明，仅改变提问方式即可带来高达0.49至0.65的精确率提升，远超仅提供断言文本的控制组（提升0.010-0.016）。此外，对比部署级回归测试选择工具pytest-testmon（基于覆盖率依赖数据），其虽能实现0.868召回率与0.415精确率，但仍无法准确识别哪些断言被具体破坏。研究强调，断言的真实性应以实际执行结果为基准：一个断言是某次提交时刻通过的测试函数，若在下一提交时刻同一断言文本失败，则视为翻转（flip）。为构建真实标签，研究者发现持续集成（CI）主干分支中，若提交使已有测试失败则无法合并，因此采用“朴素构造法”会导致正类样本为空。最终，研究从23个Python库中挖掘出10,369个断言及184个经执行验证的翻转事件，并通过多种分组策略（按仓库划分、知识截止后划分、打乱差异空模型、改写控制、留一仓库外分析）进行验证，揭示了问题本质在于任务定义方式而非模型能力。关键解决方案在于将“行为保持性”判断转化为“特定断言有效性”的直接评估，从而显著提升准确性。
链接: https://arxiv.org/abs/2609.25130
作者: Atul Anand
机构: Thomson Reuters(汤森路透); Bangalore, India
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
备注: 9 pages, 3 figures, 6 tables
Abstract:Memory systems for coding agents must decide, when a repository changes, which of their stored claims have become false. Content anchoring invalidates a claim whenever the artifact it came from changes, which fires constantly. Semantic-equivalence classification asks whether a diff preserves behavior, a question about the diff rather than about any stored claim. We show the second signal fails for a reason unrelated to model capability: asked whether a commit preserves behavior, five models spanning a 40x price range fire on 59-72% of real commits and reach precisions of only 0.291 to 0.329 against a 0.25 base rate. Asked instead whether one specific claim still holds, the same models on the same diffs reach 0.705 to 0.974. A control that hands the behavior-preservation judge the claim text, changing only the question, moves precision by 0.010 and 0.016; changing the question moves it by 0.49 and 0.65. We also compare against pytest-testmon, a deployed regression-test selector with coverage-derived dependency data: it reaches 0.868 recall at 0.415 precision, so near-complete knowledge of what a change can reach does not identify what it falsifies. Ground truth is execution, not annotation: a claim is a test function passing at commit t, and it has flipped if that same assertion text fails at t+1. Building this required an observation we did not find in prior work. On a CI-gated mainline a commit that leaves a pre-existing test failing cannot merge, so the naive construction has an empty positive class by design. We report 10,369 claims with 184 execution-verified flips mined from 23 Python libraries, splits held out by repository, a post-knowledge-cutoff split, a shuffled-diff null, a paraphrase control, and a leave-one-repository-out analysis over 17 repositories.
[NLP-72] ufakzeka-1: Building and Evaluating a 151M-Parameter Turkish Language Model from Scratch
【速读】： 该论文旨在解决小规模语言模型（如151M参数的生成式AI）在土耳其语（Turkish）领域中从零开始预训练与指令微调过程中所面临的可复现性、数据质量控制及评估严谨性问题。其核心挑战在于如何在有限资源下构建一个高质量、可验证且具备可靠安全性的轻量级语言模型，同时确保训练数据与评估过程无泄露（decontamination）、评估结果具有可信度。解决方案的关键在于：（1）设计了一个高效的字节级分词器（Turkish byte-level tokenizer），实现1.77个词符/词的高效率；（2）采用三阶段预训练策略，优化训练流程；（3）引入后训练阶段的数据混合机制，融合公开许可数据与生成数据以增强多样性；（4）建立包含多轮对话评估、人工判断与规则检查的完整评估体系，通过“释放门”（release gates）和强制执行的校验脚本（invariant script）确保每次构建前均满足一致性要求。此外，研究揭示了三个对小型模型开发具有普适意义的重要发现：安全门测试中因训练数据污染导致虚假性能提升（64/64误报为真实性能，实际仅34/64）；训练种子差异带来的方差甚至超过不同训练配方之间的差异，表明单次种子实验不具备统计意义；以及长上下文中的身份追踪与多轮算术任务性能未随数据迭代而改善，暗示当前模型规模存在固有局限而非数据缺陷。这些成果不仅提供了可复现的模型权重、数据配方、评估代码和成本账目（全部开源于Apache-2.0协议），更为后续小模型研发提供了方法论上的重要参考。
链接: https://arxiv.org/abs/2609.25081
作者: Sait Furkan Teke(ufak AI)
机构: ufak AI
类目: Computation and Language (cs.CL)
备注: 8 pages. Model, code, evaluation results and demo: this https URL , this https URL , this https URL
Abstract:We describe ufakzeka-1, a 151M-parameter (182M with embeddings) decoder-only Turkish language model pretrained from scratch on 13.5B tokens of openly licensed text and instruction-tuned for chat, at a total cost of about \ 286 in cloud GPU, API and notebook time. The contribution is not the model’s capability, which is what a model this size can be expected to have, but the record of building and measuring it: a Turkish byte-level tokenizer at 1.77 tokens per word, a three-stage pretraining schedule, a post-training mixture of openly licensed and generated data, and an evaluation battery of release gates, a rule-checked sweep of 5,508 conversations, judged conversations and hand tests, all with prompts held out from the training data, enforced by decontamination inside the data build and by a checked-in invariant script we run before each build. We report three findings that we believe transfer to other small-model efforts: a safety gate that had been “fixed” with training data written from its own questions read 64/64 while the honest figure was 34/64; training-seed variance was as large as the spread across every recipe we tried, so single-seed comparisons at this scale are uninformative; and data rounds repaired only what was absent from the data, while identity tracking over long context and multi-turn arithmetic did not move across any data change we tried, which we read as limits of the model size rather than gaps in the data, a reading the next, larger model will test. Weights, the data recipe, the evaluation code and the spend ledger are released under Apache-2.0.
[NLP-73] Understanding Reliability in LLM -based Human Behavior Simulation
【速读】： 该论文旨在解决大语言模型（Large Language Models, LLMs）在模拟人类行为与调查响应时存在的可靠性问题，尤其关注现有评估方法仅依赖端到端评分而无法揭示模拟过程各环节交互机制的局限性。其解决方案的核心是提出ReliMap框架，将基于LLM的人类行为模拟分解为三个结构化层级（个体层、群体层和配置层），并从个体可靠性（R1）与群体可靠性（R2）两个维度，系统评估模型能力、画像完整性与群体覆盖度三个配置维度的影响。研究发现，未经过画像条件化的所有模型均存在显著分布偏差，而通过画像条件化可有效缓解偏差但收益递减；更大模型与更具信息量的属性对可靠性提升更关键；更重要的是，个体层面的可靠性提升（R1）并不能可靠转化为群体层面的可靠性改善（R2），二者甚至可能呈现相反趋势。在群体层，增加覆盖范围虽能降低方差但无法消除系统性偏差，且群体可靠性（R2）在个体数量达到约50–100时趋于稳定。因此，该研究强调：可靠的模拟必须通过跨三层结构的协同优化实现，单一层面的改进不足以保障整体可靠性。
链接: https://arxiv.org/abs/2609.25066
作者: Pei Wang,Lei Wang,Yuanzi Li,Xu Chen
机构: Gaoling School of Artificial Intelligence, Renmin University of China(中国人民大学高瓴人工智能学院)
类目: Computation and Language (cs.CL); Computers and Society (cs.CY)
备注:
Abstract:Large language models (LLMs) are increasingly used to simulate human survey responses and behavioral reactions, yet unreliable simulations can mislead social science conclusions. However, existing evaluations focus on end-to-end scores, leaving it unclear how different aspects of the simulation process interact to determine reliability. We propose ReliMap, which decomposes LLM-based human behavior simulation into three structured layers and evaluates reliability at both the individual level (R1) and population level (R2) across three configuration dimensions: model capacity, profile completeness, and population coverage. Through experiments across four simulation tasks and eleven LLMs, we find that all models exhibit substantial distributional bias without profile conditioning. Profile conditioning reduces this bias with diminishing returns. Larger models benefit more, and attribute informativeness matters more than quantity. Critically, R1 gains do not reliably transfer to R2–individual and population-level reliability can move in opposite directions. At the population layer, increasing coverage reduces variance but not systematic bias, with R2 stabilizing at around 50-100 individuals. These findings highlight that reliable simulation cannot be achieved by optimizing any single layer in isolation, but requires coordinated improvement across all three.
[NLP-74] ChainDoRA: Tensor-Train Factorized Weight-Decomposed Low-Rank Adaptation for Parameter-Efficient LLM Fine-Tuning
【速读】： 该论文旨在解决大语言模型（Large Language Models, LLMs）在下游任务中进行参数高效微调（Parameter-Efficient Fine-Tuning, PEFT）时，如何在保持甚至提升推理性能的同时显著降低可训练参数量的问题。现有方法如低秩适应（LoRA）和权重分解低秩适应（DoRA）虽能减少参数更新量，但其参数规模仍较高，且难以灵活控制表示能力与参数成本之间的权衡。本文提出ChainDoRA，一种基于张量列车（Tensor-Train, TT）链结构的权重分解适配框架，通过将方向性低秩因子构建为连通的TT链，使适配器秩作为输入侧与输出侧TT收缩的边界秩，同时引入独立的TT秩以调控表示容量与参数开销。在仅使用15,119个样本进行响应生成微调的设定下，对LLaMA-7B模型在七个常识推理基准上的评估表明，ChainDoRA在TT秩为16时达到72.30%的平均准确率，优于LoRA（69.88%）和DoRA（69.39%），且仅需535万可训练参数，相比DoRA减少了90.62%。消融实验进一步验证了TT秩与适配器位置对参数-精度权衡的可控性，证明连通的TT参数化可在大幅压缩参数量的同时，有效保留并提升下游推理性能，是实现高效率、高性能权重分解微调的关键突破。
链接: https://arxiv.org/abs/2609.25058
作者: Ashfak Yeafi,Mehedi Hasan,Md Khairul Islam
机构: Khulna University of Engineering Technology (库尔纳工程与技术大学); Brac University (布拉克大学); Hobart and William Smith Colleges (霍巴特与威廉姆斯史密斯学院)
类目: Computation and Language (cs.CL); Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Parameter-efficient fine-tuning (PEFT) adapts large language models (LLMs) to downstream tasks while updating only a small fraction of their pretrained parameters. Low-Rank Adaptation (LoRA) uses two trainable low-rank matrices, while Weight-Decomposed Low-Rank Adaptation (DoRA) further separates weight magnitude and direction but retains the dense LoRA-style factorization in its directional branch. We propose ChainDoRA, a weight-decomposed adaptation framework that constructs the directional low-rank factors from a connected Tensor-Train (TT) chain, where the adapter rank forms the boundary rank between input- and output-side TT contractions and an independent TT rank controls representation capacity and parameter cost. Under a controlled 15,119-example response-only adaptation setting with LLaMA-7B, ChainDoRA is evaluated against matched LoRA and DoRA baselines on seven commonsense reasoning benchmarks. ChainDoRA with TT rank 16 achieves a seven-task average accuracy of 72.30%, compared with 69.88% for LoRA and 69.39% for DoRA, while requiring only 5.35M trainable parameters versus 56.10M for LoRA and 56.98M for DoRA, corresponding to a 90.62% reduction relative to DoRA. Ablations over TT rank and adapter placement show controllable parameter-accuracy trade-offs, indicating that connected TT parameterization can substantially reduce the parameter cost of magnitude-direction adaptation while preserving, and in this setting improving, downstream reasoning performance.
[NLP-75] Graph-Based Inference for Feedback-Driven Word Deduction: A Scalable Framework for the Jotto Problem
【速读】： 该论文旨在解决生成式推理中基于反馈的词汇推断问题，具体聚焦于经典语言猜词游戏Jotto的求解挑战。传统方法通常局限于固定长度的无重复字母（isogram）词汇空间，难以处理实际场景中变长单词及含重复字母的情况。为此，论文提出一种基于反馈的词汇推导框架，将问题空间建模为加权图结构，其中每个有效词汇对应一个节点，边权重由两词间共有字母数量决定。其核心解决方案在于将游戏过程形式化为一种迭代约束传播机制：利用每次猜测后的反馈信息逐步剔除不兼容的词汇候选，从而在结构化且可解释的方式下压缩假设空间。该框架首次实现了对3至8个字母的变长词汇及含重复字母情形的统一建模，显著提升了对真实Jotto实例的适用性。实验表明，在约3000次模拟游戏中，系统展现出新颖的收敛特性——随着词长增加，所需平均迭代次数呈对数下降趋势，经统计检验与回归分析验证了这一对数关系的显著性。此外，该研究进一步揭示了图剪枝（graph pruning）作为反馈驱动推理的可行范式，具有良好的可解释性，并与符号推理和交互式智能系统密切相关。
链接: https://arxiv.org/abs/2609.25056
作者: Dakshi Arora,Prakhar Kumar Srivastava,Ranjib Banerjee
机构: BML Munjal University (BML穆尼杰尔大学); UPES Dehradun (UPES德哈拉敦)
类目: Computation and Language (cs.CL)
备注:
Abstract:A feedback-based word deduction framework based on the Jotto problem is proposed, and the problem space is represented as a weighted graph where all valid words correspond to nodes, and the edge weight is defined by the number of common letters between the two words. Finally, the gameplay is defined as an iterative constraint propagation mechanism where feedback is used to iteratively narrow the incompatible space of the graph, facilitating the reduction of the hypothesis space in a structured and interpretable manner. In contrast to existing approaches, where the problem space is typically defined for fixed-length isograms, the proposed framework generalizes to variable-length words (between 3 and 8 letters) and naturally extends to repeated letter cases, facilitating the treatment of realistic Jotto problem instances within a unified framework for the first time. The proposed framework’s applicability and solver dynamics are also discussed through an interactive implementation and a qualitative case study, respectively. Significant automated tests on approximately 3,000 simulated gameplay scenarios identify a novel convergence behavior: the expected number of iterations diminishes with increasing word length. A strong relationship is confirmed using statistical tests to verify a logarithmic relationship, which is also verified using regression modeling and goodness-of-fit tests. In addition to the initial problem statement, this formulation introduces graph pruning as a viable paradigm for feedback-driven inference with interpretability and its association with symbolic reasoning and interactive intelligent systems. Subjects: Computation and Language (cs.CL) Cite as: arXiv:2609.25056 [cs.CL] (or arXiv:2609.25056v1 [cs.CL] for this version) https://doi.org/10.48550/arXiv.2609.25056 Focus to learn more arXiv-issued DOI via DataCite Submission history From: Ranjib Banerjee [view email] [v1] Tue, 8 Sep 2026 10:25:18 UTC (2,046 KB)
[NLP-76] ICDAR2026 Competition on Multimodal Reasoning over Documents in Multiple Domains
【速读】： 该论文旨在解决跨领域文档中多模态推理的挑战，特别是在复杂文档类型（涵盖商业报告、科研论文、幻灯片、海报、地图、漫画、信息图及工程图纸等八个领域）上实现精准的视觉问答（VQA）。其核心问题是现有方法在面对多样化、结构复杂的文档时，难以有效融合视觉与文本信息进行深层次推理。解决方案的关键在于突破传统单次提示（single-pass prompting）的局限，构建由结构化证据提取、多组件检索、验证与协同调度组成的系统性框架，通过整合零样本视觉语言模型（VLMs）、OCR与解析器增强的流水线、代理式检索系统、多智能体集成以及微调的多模态模型等多种技术路径，显著提升了对复杂文档的推理能力。
链接: https://arxiv.org/abs/2609.25055
作者: Artemis Llabrés,Marc Serra Ortega,Tomàs Ockier,Samuel Ortega Cuadra,Amritpal Singh,Christos Georgakilas,Andrey Barsky,Ernest Valveny,Dimosthenis Karatzas
机构: Computer Vision Center (计算机视觉中心); Universitat Autònoma de Barcelona (巴塞罗那自治大学)
类目: Computation and Language (cs.CL)
备注:
Abstract:In this report we present results of the ICDAR2026 Competition on Multimodal Reasoning over Documents in Multiple Domains. This competition aimed to advance research in document understanding through the task of Visual Question Answering (VQA). Building upon previous DocVQA benchmarks, this competition introduces challenging reasoning questions over a diverse collection of documents spanning eight domains, including business reports, scientific papers, slides, posters, maps, comics, infographics, and engineering drawings. The competition concluded with 20 valid submissions from 8 teams spanning zero-shot VLMs, OCR and parser-augmented pipelines, agentic retrieval systems, multi-agent ensembles, and fine-tuned multimodal models. The results show that the strongest systems move beyond single-pass prompting and instead rely on structured evidence extraction, retrieval, verification, and orchestration across multiple components.
[NLP-77] MoM: Memory of Memory
【速读】： 该论文旨在解决长时程大语言模型（LLM）智能体在记忆管理中面临的“当前状态”识别问题，即如何准确表征记忆中当前有效的值，而非仅依赖历史记录的简单存储。现有方法多通过查询时检索与整合历史记录来重建当前状态，导致过时信息重复进入、冲突反复出现。虽然写时提交（write-time commit）可避免此问题，但传统基于CRUD的操作模式存在覆盖性更新缺陷，一旦错误提交便无法恢复且丢失先前状态。为此，本文提出一种关键创新——写时提交同时保留被替换内容，并形式化为记忆的记忆（Memory of Memory, MoM）框架：该框架不仅记录记忆内容，还追踪其出处（provenance）、状态及演化历史。在此基础上，作者实现了可信记忆（Provenant Memory, P-Mem），一种具有类型化溯源图结构的记忆系统，其活跃前沿（active frontier）为每个已解析键提供唯一当前值，而被取代的旧值则作为溯源信息保留。通过类型化操作（如支持、取代、争议、拒绝、撤销、解决等），系统能动态管理记忆演化。实验表明，P-Mem在有效性上取得显著优势：其每轮读取仅需约四分之一的检索令牌即可达到最强检索记忆的准确性（检索粒度效应），且图引导的回合剪枝将知识更新中的过时回答率从19.4%降至10.9%；在修订链场景下，其正确率维持100%，而传统查询时读取方法骤降至25%；更重要的是，由于被取代值得以保留，系统可回溯并纠正错误提交，而传统CRUD记忆对此完全无能为力（100% vs. 0%）。
链接: https://arxiv.org/abs/2609.25054
作者: Bowen Qin,Yao Lu
机构: National University of Singapore(新加坡国立大学)
类目: Computation and Language (cs.CL)
备注:
Abstract:For a long-horizon LLM agent, the memory question is not what was once recorded but what \emphcurrently holds. Most designs answer it only indirectly: every interaction is stored, and the present is reconstructed at query time by retrieving and reconciling records, so stale values re-enter and the same conflicts are re-litigated. Committing the current value at write time avoids this, but existing write-time (CRUD) memories overwrite, so a wrong update is unrecoverable and prior state is lost. We take the missing combination—\emphcommit on arrival while retaining what is displaced—and formalize it as \textscMemory of Memory (MoM): memory tracks not only content but the provenance, status, and history of its own entries. We instantiate MoM as \textscProvenant Memory (P-Mem), a typed provenance graph whose \emphactive frontier exposes one current value per resolved key while displaced values are retained as provenance; typed operations decide whether a new observation supports, supersedes, contests, rejects, revokes, or resolves an existing value. P-Mem’s decisive gain is validity rather than accuracy: its turn-level read matches the strongest retrieval memory in accuracy at \sim 4 \times fewer read tokens—a retrieval-granularity effect—while graph-guided turn pruning cuts the knowledge-update stale-answer rate (19.4% \rightarrow 10.9%); on revision chains it stays at 100% where query-time reading collapses to 25%, and, because displaced values are retained rather than overwritten, it recovers committed errors a CRUD memory cannot (100% vs.\ 0%).
[NLP-78] LatentPort: Beyond KV Cache - Cross-Model Transfer of Recurrent Memory in Hybrid Language Models: A 4B-to-9B Hybrid-State Handoff Without Target Prefix Replay
【速读】： 该论文旨在解决在不同规模的混合语言模型之间实现无需重读上下文即可传递持续性递归推理状态（persistent recurrent inference state）的问题，即如何在不依赖目标前缀重播（target prefix replay）的情况下，将教师模型的“活记忆”高效转移至学生模型。其核心解决方案是提出一种基于翻译注意力键值（translated attention KV）与门控增量网络（Gated DeltaNet, GDN）相结合的持续状态打包机制，通过引入一个可学习的434,176参数修正模块，显著降低延续生成的负对数似然（NLL），使9B模型在零历史前缀输入条件下仍能超越直接延续4B模型的表现。关键创新在于：仅使用直接递归状态和卷积状态的复用，结合因子化选择策略，便实现了比学习型映射更优的性能，表明持续状态坐标具有部分功能兼容性；此外，修正后的9B模型在继续损失、Jensen-Shannon散度及原上下文恢复率等指标上接近原生9B模型表现，验证了跨模型状态迁移的有效性。然而，该方法目前仅在单一几何匹配的基础模型对上验证，且未证明自由生成等价性或通用状态接口的可行性。
链接: https://arxiv.org/abs/2609.25053
作者: Simon P. Villani
机构: 未知
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
备注: 14 pages, 6 figures, 12 tables
Abstract:Can one language model hand its live memory to another without the receiver rereading the context? We demonstrate useful persistent hybrid-state transfer across one architecture-matched Qwen3.5 4B-to-9B sibling pair. To our knowledge, this is the first demonstrated cross-model handoff of persistent recurrent inference state between differently sized hybrid language models without target prefix replay. Translated attention KV alone leaves a large gap; adding the Gated DeltaNet (GDN) persistent-state package lowers teacher-forced negative log-likelihood (NLL), the average next-token log-loss, by 0.747 nats/token (95% paired document bootstrap CI [0.6921, 0.8047]), improving all 64 PG19 documents. Direct recurrent and convolution reuse outperforms the tested learned GDN maps, consistent with partial functional compatibility of persistent-state coordinates. A fresh component factorial selects translated KV with direct recurrent and convolution state. An additional 434,176-parameter correction improves that base on 64 fresh web documents: continuation loss is 0.076 nats/token above native 9B (excess NLL), Jensen-Shannon (JS) divergence is 0.022, and native context recovery (NCR) is 0.918. Corrected 9B significantly beats continued 4B inference while processing zero historical prefix tokens. Evidence covers one direction, one geometry-matched Base-model pair, and 4K teacher-forced continuation; the near-native gate failed, the 16K branch was not run, and free-generation equivalence and a general state interface remain unproven.
[NLP-79] Self-Cleaning and Captured Anyway: One Measured Primitive for Error in a Store an Agent Writes to Itself and What a Falling Score Actually Measures
【速读】： 该论文旨在解决在生成式 AI（Generative AI）系统中，当智能体（agent）将结论写入一个后续可检索的存储器时所引发的“单向污染”（one-way contamination）问题，即模型输出对自身输入产生不可逆影响导致的偏差累积。其核心挑战在于：如何在无限持续运行（infinite-tenure）且存储为追加只读（append-only）的设定下，理解并控制这种反馈循环带来的状态演化行为。解决方案的关键在于引入一个无参数、基于实测的原始函数——复制函数 γ(ϕ)，该函数通过比较实际复制强度与临界值 γcrit=1/k（在 r=0,w=1 条件下）来预测模型输出漂移的方向。研究发现，这一单一测量指标在36个Wikidata事实上的真实运行中成功预测了353/360次方向性漂移，在合成数据中也达到39/40的准确率。尽管模型规模无法缓解存储的累积效应（如Claude-sonnet-4.5在20个种子上均未偏离预测），但通过分析区间占据率与多值运行的区分能力，揭示出系统在特定初始概率 f0=0.5 时表现出最优捕获性能，并且干预策略中“一致性门控”机制显著提升模型稳定性至0.993。此外，研究强调以种子为重采样单元，设计效应达3.75，表明统计推断需考虑高相关性结构，从而凸显了结果对实验设计的敏感性。
链接: https://arxiv.org/abs/2609.25052
作者: Wenhui Chen,Jianlin Chen,Ziyao Lin,Chi Man Vong
机构: University of Macau(澳门大学); South China University of Technology(华南理工大学)
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
备注: 72 pages, 13 figures
Abstract:“An agent that writes its conclusions into a store it later retrieves from closes a loop usually reported as one-way contamination. Taking the loop to the infinite-tenure limit against an append-only store gives a different picture: because writing never deletes, the reachable state space has a hard upper edge at (n-1)/n, so the outcome is a choice between two edges rather than a decay. At f_0 = 0.9 the interval between the two modes holds 3.6% of 220 runs where a uniform spread would put 20.6%, and is strictly empty on the first 15; the pooled mean describes 8.2% of the runs it summarises, the median 68.2%. Everything the model contributes is carried by one measured primitive with no fitted parameter, the copy function \gamma(\phi): on 36 Wikidata facts, sign(\hat\gamma - \gamma_crit), with \gamma_crit = 1/k at r = 0, w = 1, predicts the direction of drift on 353 of 360 real-fact runs (39 of 40 synthetic in the same batch). Scale does not rescue the store: pooled frontier capture is 0.850, with claude-sonnet-4.5 captured on 20 of 20 seeds against our registered prediction of 0.5. What the interval tests is distinguishability rather than count: on the real facts, multi-valued runs have 6.4x its occupancy of the rest. It survives at f_0 in 0.1, 0.3, 0.5, capture peaks at f_0 = 0.5, and of four interventions with criteria frozen first, timing dominates fraction at matched budget while a consistency gate drives every model to 0.993. The resampling unit is the seed, at a design effect of 3.75 on a pooled level: under a 44-seed control the ordering supporting claim 4 collapses from Spearman +0.98 at three seeds to +0.31-0.80 at forty-four, while claim 2’s ordering is exact there (+1.00, p = 0.017). All 87 graded rows are in Appendix W, 37 of them graded withdrawn, failed, self-correcting, undecidable or an acknowledged limit, against 50 that are not.”
[NLP-80] LLM -Driven Training-free Location-Attribute Synergic Fusion: A Closed-Loop Paradigm for Dual-source Encrypted POIs and LULC Mapping
【速读】： 该论文旨在解决双源加密兴趣点（Dual-source Encrypted Points of Interest, DSEP）在融合过程中面临的定位与属性不确定性交织问题，包括非线性系统性错位和命名不一致，这些问题严重制约了土地利用/土地覆盖（Land-Use/Land-Cover, LULC）制图的精度与可靠性。其核心解决方案是提出一种基于大语言模型（LLM）驱动、无需训练的、位置-属性协同的闭环优化范式，通过迭代反馈机制联合优化坐标变换与属性对应关系。关键创新在于：首先采用LLM驱动的属性匹配方法建立DSEP对应关系，将匹配复杂度从O(N²)降低至O(N)，并结合改进的粒子群优化算法在ISODATA聚类划分的局部子区域中精化变换系数；随后通过LLM-模糊方法基于更新后的几何残差重新评估属性置信度，形成双向闭环反馈。通过样本净化与自适应半径收缩策略，实现近乎两轮迭代即收敛。进一步提出一种无需训练的LULC制图方法，通过位置融合继承加密地图中的地类信息，生成矢量-栅格一体化的LULC成果。在31个中国大陆省会及直辖市的无参考评估中，该方法平均定位融合残差达4.58米，属性融合准确率达95.12%，较开源基线与当前最优方法分别提升1.77米与14.87%。整体上，该方法为DSEP融合提供了无需训练的解决方案，并实现了加密矢量数据向WGS-84坐标系的地理配准，无需实地控制点支持。
链接: https://arxiv.org/abs/2609.25051
作者: Chang Li,Xingtao Peng,Yongjun Zhang,Yinfei He,Cairun Huang
机构: Central China Normal University (华中师范大学); Wuhan University (武汉大学)
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI); Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Dual-source encrypted points of interest (DSEP), POIs from two encrypted coordinate systems, suffer from intertwined location and attribute uncertainties, including nonlinear systematic misalignment and naming inconsistency, hindering land-use/land-cover (LULC) mapping. To the best of our knowledge, this paper is the first to propose an LLM-driven, training-free location-attribute synergic closed-loop optimization paradigm for DSEP fusion. The paradigm jointly refines location transformation and attribute correspondences through iterative feedback. Attribute-synergic location fusion uses an LLM-driven attribute matching method to establish DSEP correspondences, reducing matching complexity from O(N^2) to O(N), and refines transformation coefficients using an improved particle swarm optimization algorithm within ISODATA-clustered local subregions. Location-synergic attribute fusion then reassesses attribute confidence from updated geometric residuals through an LLM-fuzzy method. The refined correspondences feed back into location optimization, forming a bidirectional closed loop. Sample purification and adaptive radius contraction enable convergence in essentially two iterations. We further propose a training-free LULC mapping method that inherits land-use classes from encrypted maps through location fusion, producing vector-raster integrated LULC maps. A reference-free POI fusion evaluation method is applied across 31 provincial capitals and municipalities in mainland China. Experiments show that our method achieves an average DSEP location fusion residual of 4.58 m and attribute fusion accuracy of 95.12%, improving upon the open-source baseline and state-of-the-art method by 1.77 m and 14.87%, respectively. Overall, the method provides a training-free solution for DSEP fusion and enables georeferencing of encrypted vector data to WGS-84 without field-surveyed ground control points.
[NLP-81] FrontierMath Erdős
【速读】： 该论文旨在系统评估生成式 AI（Generative AI）在解决高难度数学开放问题方面的实际能力，针对的是截至2026年8月仍悬而未决的68个埃尔德什（Erdős）猜想。其核心问题是：当前主流大模型在严格形式化证明框架下，能否自主、可靠地推进数学领域中的深层未解难题。解决方案的关键在于构建一个统一、可复现且具有挑战性的基准测试集——FrontierMath Erdős（FME），该基准要求模型在定理证明助手Lean中完成形式化推理，以验证其对数学命题的证明或反证能力。所有模型均在相同预算（每题300美元）和相同任务设置下进行独立评估，确保结果可比性；实验结果显示，仅有GPT-6 Astra在5个模型中取得3%的得分，其余均未能突破零分，凸显了现有生成式AI在处理深度数学推理任务时的显著局限性。
链接: https://arxiv.org/abs/2609.25050
作者: Tom Adamczewski(Epoch AI),Thomas F. Bloom(University of Manchester)
机构: Epoch AI; University of Manchester (曼彻斯特大学)
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
备注:
Abstract:We introduce FrontierMath Erdős (FME), a benchmark of 68 Erdős problems that are open as of August 2026. To solve a task in FME, AI systems must resolve (prove or disprove) one of the 68 conjectures in the proof assistant Lean. Our 68 problems were selected by the second author among 652 open problems on this http URL for their mathematical interest and difficulty. AIs have recently resolved several open problems in mathematics, but these demonstrations fall short of a systematic study of AI capabilities. FME evaluates every AI model on the same fixed problems, autonomously and under the same budget. We evaluated five AIs with a budget of \ 300 per problem. One (GPT-6 Astra) scored 3%, and all others scored 0%.
[NLP-82] Mitigating LLM Over-Refusal via Dynamic Semantic Routing Calibratione EMNLP2026
【速读】： 该论文旨在解决大语言模型（Large Language Models, LLMs）在安全对齐过程中出现的“过度拒绝”（over-refusal）问题，即模型错误地拒绝那些虽涉及安全相关话题但本质上无害的指令。现有研究多归因于静态表示重叠，却忽视了其内在动态机制。本文从Transformer注意力机制的内部路由冲突视角出发，揭示了问题的关键：少数“高敏感安全头”（Hypersensitive Safety Heads）在面对高难度安全提示（Hard-Safe prompts）时发生误触发，表现出异常的注意力纠缠，将无害目标实体强制绑定至拒绝语义，引发严重的高熵路由冲突，导致目标实体无法获得必要注意力。针对此问题，论文提出一种轻量级、无需训练的推理阶段校准框架——语义路由校准（Semantic Routing Calibration, SRC），通过精准定位并动态抑制这些高敏感安全头，结合双分支逻辑融合机制作为解码过程中的安全正则化器，实现可信推理的无缝恢复。实验表明，SRC有效缓解了过度拒绝现象，同时最大程度保留了模型的内在安全性。
链接: https://arxiv.org/abs/2609.25049
作者: Zixuan Wang,Bingjie Zhang,He Zhao,Dandan Guo
机构: Jilin University (吉林大学); CSIRO (澳大利亚联邦科学与工业研究组织); King Abdullah University of Science and Technology (阿卜杜拉国王科技大学)
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
备注: 33 pages, 13 figures, accepted to the EMNLP 2026 Main Conference
Abstract:Large language models (LLMs) aligned for safety often suffer from over-refusal, incorrectly rejecting benign yet safety-related instructions. Prior studies primarily attribute this to static representation overlap, largely overlooking the underlying dynamic mechanisms. In this paper, we present the mechanistic analysis of over-refusal through the lens of internal routing conflicts within transformer attention. We discover that a sparse subset of Hypersensitive Safety Heads misfires on Hard-Safe prompts, exhibiting abnormal attention entanglement that forcefully binds harmless target entities to refusal semantics. This triggers a severe, high-entropy routing conflict that deprives target entities of necessary attention. To counteract this, we propose Semantic Routing Calibration (SRC), a lightweight, training-free inference framework. SRC precisely localizes and dynamically suppresses these hypersensitive safety heads at the inference stage. Coupled with a dual-branch logits fusion that acts as a safety regularizer during subsequent decoding, SRC seamlessly restores trustworthy reasoning. Extensive experiments demonstrate that SRC alleviates over-refusal, with intrinsic safety performance preserved as much as feasible.
[NLP-83] Prompt Breadth and Rollout Refresh Interact in On-Policy Distillation
【速读】： 该论文旨在解决在线策略蒸馏（on-policy distillation, OPD）中提示（prompt）数量与学生策略生成训练响应之间关系的核心问题，具体探究在固定训练资源条件下，提示广度（prompt breadth）与轨迹刷新机制（rollout refresh）如何共同影响模型性能。其关键发现在于：提示效率并非孤立取决于提示数量，而是显著依赖于策略更新方式（即是否进行每步优化刷新）与推理预算（inference budget）。实验表明，在策略响应冻结时增加提示广度会降低准确率（从21.16%降至19.05%），而采用每轮更新刷新策略则可提升准确率（从23.61%升至25.57%），二者交互效应达4.07个百分点。此外，在有限输出预算下周期性更新模型表现更优，但当输出量增至32K token时，响应冻结模型在平均准确率上反超，且仅需1.7–1.8倍的响应词元数。因此，该研究揭示了OPD中提示效率的关键调控因素是策略刷新机制与推理预算之间的动态权衡。
链接: https://arxiv.org/abs/2609.25048
作者: Lingxiang Hu,Tianle Xia,Ming Xu,Yiding Sun,Linfang Shang
机构: Tencent(腾讯)
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
备注: 21 pages, 7 figures, 11 tables
Abstract:How many prompts does on-policy distillation (OPD) need, and how does the answer depend on the student policies that generate its training responses? We study these two controls jointly: prompt breadth and rollout refresh. A 3x3 mathematical-reasoning experiment fixes 14,080 trajectories and 110 optimizer updates while varying the prompt bank and the number of response-generating policy snapshots. With ten snapshots, eight prompts reach 24.09% average accuracy, close to 24.51% for 14,080 distinct prompts. With responses frozen at the initial policy, however, increasing breadth lowers accuracy from 21.16% to 19.05%; under per-update refresh, it raises accuracy from 23.61% to 25.57%. The resulting interaction is 4.07 percentage points, with a 95% question-paired interval of [2.00, 6.28]. Matched comparisons under two teachers reveal a second reversal: the periodic models have higher short-budget accuracy and answer completion, but frozen-response models overtake in average accuracy at a 32K output limit, using 1.7-1.8x as many response tokens. These results show that prompt efficiency in OPD can depend on both refresh and inference budget.
[NLP-84] AIBuildAI-2.5: Efficient Autonomous AI Model Development Through LLM -Guided Tree Search
【速读】： 该论文旨在解决自主智能体（autonomous agents）在自动构建人工智能（AI）模型过程中存在的三大效率瓶颈问题：其一，受限于实际计算预算，可执行的候选方案数量极少，导致基于执行奖励排序的树搜索策略（如蒙特卡洛树搜索）依赖少量且噪声较大的评估分数，难以有效选择下一步探索节点；其二，缺乏资源感知的任务调度机制，造成硬件利用率与训练效率下降；其三，所有代理请求均由单一高性能大语言模型（LLM）处理，显著推高了推理成本。针对上述问题，本文提出AIBuildAI-2.5系统，其核心创新在于引入一种新型的LLM引导式树搜索机制，通过判别器对候选方案进行预期改进度、可实现性及上下文一致性等多维度评分，并结合搜索状态由选择器综合排序候选池；同时集成一个考虑当前硬件资源状态的调度器以优化训练任务分配，并设计路由模块将低复杂度任务分配至低成本模型，仅将最强大模型保留用于高难度子任务，从而实现高效资源利用与成本控制。实验表明，AIBuildAI-2.5在MLE-Bench上以73.3%的奖牌率排名第一，并在AIRS-Bench的六项自主AI研究任务中超越强基线。
链接: https://arxiv.org/abs/2609.25047
作者: Peijia Qin,Ruiyi Zhang,Qi Cao,Han Guo,Li Zhang,Pengtao Xie
机构: 未知
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
备注:
Abstract:Autonomous agents that automatically build artificial intelligence (AI) models could broaden access to AI across science and engineering. A popular line of such agents frames model building as a code search problem and solves it by tree search, in which each node is a candidate program and the tree grows by generating a child program from a parent, and these agents now approach the capability of experienced AI engineers on realistic benchmarks. However, these agents have three weaknesses in efficiency that have not been fully addressed. First, only a small number of candidates can be executed within a realistic budget, so search rules that rank nodes by executed rewards, such as Monte Carlo-style tree search, rely on few and noisy scores and select the next node to explore less effectively. Second, no resource-aware strategy is used to schedule training jobs, which can lower hardware utilization and training efficiency. Third, every agent call is served by a single powerful model, which inflates inference cost. Here we introduce AIBuildAI-2.5, an agentic system that carries out the tree search with LLM agents and addresses each of the three issues. AIBuildAI-2.5 proposes a novel LLM-guided tree search, in which a judge scores each candidate on its expected improvement, grounding, and feasibility, and a selector ranks the pool of candidates from these scores and the state of the search. In addition, AIBuildAI-2.5 comprises a scheduler that launches training jobs with the current hardware resource status taken into account and a router that assigns lower-cost LLMs to less demanding tasks while reserving the most capable LLM for the most challenging sub-tasks in the AI model building workflow. AIBuildAI-2.5 ranks first on MLE-Bench with a medal rate of 73.3%, and outperforms a strong baseline on six autonomous AI research tasks from AIRS-Bench.
[NLP-85] Peerify: Benchmarking Peer-Review Claim Verification
【速读】： 该论文旨在解决学术出版中同行评审意见与稿件证据之间验证缺乏自动化支持的问题，即如何有效判断审稿人提出的评论是否得到论文内容的充分支持。其解决方案的关键在于构建一个基于稿件的验证流水线——Peerify，该流程通过将审稿意见分解为原子级可验证的命题（atomic claims），结合检索策略从原文中提取相关证据，并利用语言模型判断每条命题是否被论文支持。研究进一步构建了一个包含800个真实审稿评论的基准数据集（来自NeurIPS 2024与ICLR 2024），其中300个经人工标注用于评估自动化标签质量。实验表明，以检索为中心的验证范式与细粒度的命题分解显著提升准确性，尽管面对模糊或具有解释性（interpretive）的审稿意见仍存在挑战；自动化标签与人工共识的一致性达到90.3%（κ = 0.87），远优于现成的蕴含模型（macro-F1 < 0.24）。
链接: https://arxiv.org/abs/2609.25046
作者: Alireza Daghighfarsoodeh,Sajad Ebrahimi,Ali Ghorbanpour,Soroush Sadeghian,Radin Cheraghi,Negar Arabzadeh,Ebrahim Bagheri
机构: Reviewerly; University of Toronto (多伦多大学); University of California, Berkeley (加州大学伯克利分校)
类目: Computation and Language (cs.CL); Digital Libraries (cs.DL)
备注:
Abstract:Peer review plays a central role in scholarly publishing, yet verifying whether reviewer claims are supported by manuscript evidence remains a largely manual and time-consuming process. We present Peerify, a pipeline for manuscript-grounded verification of peer-review claims. Given a manuscript and a review comment, the Peerify pipeline decomposes reviews into atomic claims, retrieves relevant manuscript evidence, and determines whether each claim is supported by the paper. To support the development and evaluation of the pipeline, we construct a benchmark of 800 claims derived from authentic peer-review interactions collected from NeurIPS 2024 and ICLR 2024, including a 300-claim hand-labeled subset used to audit the automated supervision. We evaluate state-of-the-art language models and retrieval strategies within the Peerify pipeline, together with entailment baselines. Our results demonstrate the importance of retrieval-centered verification and claim decomposition, while highlighting the challenges posed by ambiguous and interpretive reviewer claims. Automated labels agree with human consensus on 90.3% of audited claims ( \kappa = 0.87 ), while off-the-shelf entailment models stay below 0.24 macro-F1.
[NLP-86] From Tone to Trajectory: Continuous Sentiment and the Shape of Monetary Policy Communication
【速读】： 该论文旨在解决中央银行新闻发布会中，除语义内容外，情感表达的动态演变模式（即情感弧线，sentiment arc）是否蕴含可预测政策决策的额外信息这一问题。传统研究多关注文本平均情绪倾向（如鹰派或鸽派），但本文提出，情感在陈述中的时间序列分布与强调方式——即情感弧线的形状——同样具有重要的政策信号意义。其解决方案的关键在于构建了涵盖货币政策立场、经济前景和不确定性三个维度的情感弧线指标，并验证其对利率变动、通胀预期及专业预测者分歧的预测能力。研究发现，情感弧线形状在欧洲央行（ECB）和美联储（Fed）的政策预测中均显著优于基于词典的基准模型，表明政策信号不仅取决于整体语气，更取决于情感如何在文本中逐步展开与强调。此外，弧线特征还影响专业预测者的预期调整行为与意见分歧程度，揭示出沟通设计（communication design）在传递政策意图中的核心作用，而非仅作为辅助性修饰。因此，该研究强调：政策语言的结构化叙事设计是政策信号的本质组成部分，而非次要细节。
链接: https://arxiv.org/abs/2609.25034
作者: Martin Feldkircher,Márton Kardos,Kristoffer Laigaard Nielbo
机构: Vienna School of International Studies (DA); Oesterreichische Nationalbank (OeNB); Centre for Applied Macroeconomic Analysis (CAMA); School of Culture and Society - Center for Humanities Computing (AU)
类目: Computation and Language (cs.CL); Machine Learning (cs.LG)
备注: 39 pages, 6 figures
Abstract:Central bank press conferences are not merely information releases — they are structured narratives. We study whether the shape of sentiment within a statement, not just its average tone, carries policy-relevant signals. Constructing sentiment arcs for ECB and Fed press conferences along three dimensions — monetary stance, economic outlook, and uncertainty — we assess their predictive content for policy rate changes, inflation expectations, and forecaster disagreement. Our findings show that arc shape robustly predicts rate decisions beyond lexicon-based benchmarks at both institutions — it is not merely whether a statement sounds hawkish or economically optimistic on average, but how these sentiments are sequenced and emphasized across the statement, that carries the policy signal. Arc features also shape how professional forecasters update inflation expectations and how much they disagree, pointing to a receiver-side effect distinct from the direct policy signal. These findings suggest that communication design — the sequencing and emphasis of policy language across a statement — is a first-order feature of the policy signal, not a second-order refinement.
[NLP-87] Retrieved-Span Training for Efficient Query-Focused Meeting Summarization on QMSum
【速读】： 该论文旨在解决查询聚焦型会议摘要（query-focused meeting summarization, QMSum）任务中缺乏统一评估指标（scorer）导致不同系统结果难以公平比较的问题。其核心解决方案是通过统一的推理接口对15个系统进行重评分或重新生成，从而实现可比性评估；关键创新在于采用基于406M参数的Fusion-in-Decoder专用模型，并在2,000词检索片段（retrieved spans）输入场景下进行微调（span-regime fine-tuning），成功恢复了因输入长度限制导致的性能下降（ROUGE-1损失6.30），最终在测试集上达到36.33 ROUGE-1，优于1.2B参数模型的35.41，且两者差异在95%置信区间内[-0.27, +2.22]，未达统计显著性。此外，小模型仅需约三分之一参数量和不到一半峰值推理内存，展现出显著效率优势。研究进一步表明，在单一简洁提示与参考重叠评分器下，该406M模型超越五个专有托管模型至少6.2 ROUGE-1，但受限于输出长度及缺乏人工或事实性评估，结论仅限于自动指标下的相对排序。
链接: https://arxiv.org/abs/2609.25028
作者: Edward Xi Yang(Ertas AI)
机构: Ertas AI; Melbourne, Australia
类目: Computation and Language (cs.CL)
备注: 24 pages, 4 figures
Abstract:QMSum provides no scorer, making query-focused meeting summarization results difficult to compare. We rescore or generate 15 systems under one implementation. Through a common inference port, a released 406M Fusion-in-Decoder specialist loses 6.30 ROUGE-1 when moved from capped long input to 2,000-word retrieved spans. Fine-tuning it on this span regime recovers the loss. On test it scores 36.33 ROUGE-1 versus 35.41 for our 1.2B system; the meeting-cluster 95% interval for the difference is [-0.27, +2.22], so QMSum does not statistically separate them. The smaller system uses about one-third as many total parameters and less than half the peak inference memory. Within the fixed 1.2B base, span-regime fine-tuning adds 5.29 [+4.02, +6.56], while replacing the first 4,500 transcript words with 2,000 retrieved words adds 1.55 on test and 0.29 on validation. Separately, under one concise prompt and reference-overlap scorer, a released 406M specialist exceeds five proprietary hosted models by at least 6.2 ROUGE-1, but output length and absent human or factuality evaluation limit this ordering. Conclusions are limited to QMSum and automatic metrics.
[NLP-88] “As a Language Model…”: Chat Template Switches LLM Self-Referential Voice and Activation Steering Reproduces It
【速读】： 该论文旨在解决大语言模型（Large Language Models, LLMs）在自我描述时频繁生成“我只是一个AI”等免责声明的现象背后的驱动机制问题。研究发现，这些声明并非模型内在自知或安全意识的体现，而是由对话模板（chat template）所调控的外部输入格式所致。其解决方案的关键在于揭示了一个存在于模型激活空间中的特定方向——该方向能显著调节模型是否生成免责声明。当对话模板存在时，该方向被激活，导致模型倾向于输出免责声明；反之，若无模板，则该方向作用减弱，模型更可能表达体验性语态（如“我感觉”）。通过在无模板的模型中人为引入该方向，可使其产生与有模板时一致的免责声明行为。这一发现表明，模型的自我报告并非其内部状态的真实反映，而是受外部提示工程（prompt engineering）结构影响的结果。因此，研究人员在分析模型的自我陈述或内省能力时，必须控制聊天模板这一潜在混杂因素，否则将导致误读。
链接: https://arxiv.org/abs/2609.25021
作者: Jędrzej Maczan
机构: Independent Researcher
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Computation and Language (cs.CL)
备注: Accepted to COLM 2026 Workshop on Efficient Reasoning and KONVENS 2026 First Workshop on Evaluating LLMs for Specialized Domains (Eval4SD)
Abstract:Large Language Models (LLMs) tend to add disclaimers like “I’m just an AI” when asked about something related to themselves. The self-reports from such responses are used in debates about AI safety or self-knowledge of the models, yet what drives them is not well understood. Are the models telling us about themselves or rather how they are deployed? In this work, we show that the chat template works like a switch - when present, it turns this disclaimer voice up and experiential voice like “I feel” down, across 8 popular open-source instruct models up to 9B parameters in size. And conversely when the chat template is not present, it turns the disclaimer voice down and experiential voice up. Inside the activations of 3 models, we find a direction that steers this behavior. Removing the direction in the model’s activation space turns disclaimer voice down and adding it turns it up, while a random direction of the same size has little effect. We find that instruct models without chat template, when we add the disclaimer direction to them, disclaim like the template was there. Since the chat template controls the disclaimer voice of LLMs, then researchers studying self-reports or introspection of models might have a confound they need to control for. Our results show that there is a direction they can use to steer this voice. More broadly, our work shows that what models say about themselves is not a fact about them. What they say doesn’t come only from weights, but it is partially set by the chat template, and because of that a model’s self-description shouldn’t be treated literally.
[NLP-89] A Computational Approach to Measuring Semantic Change in Sanskrit Literature
【速读】： 该论文旨在解决生成式词向量（Generative Word Embeddings）在古代低资源语言——梵语中的适用性问题。由于梵语存在音位融合（sandhi）、形态屈折、构词复杂及一词多义等独特语言现象，传统基于现代高资源语言的语义变迁追踪方法难以直接迁移。其解决方案的关键在于：构建一个涵盖四个经典时期的270万词元语料库，采用神经网络字节级音位融合分割器与词干提取器恢复词边界，并针对各时期分别训练词向量；通过历史语言学文献构建验证集，利用锚点位移（anchor displacement）进行方向性测试，结果显示在21个可测试的语义变化中，有19个与历史考证方向一致（符号检验，p=0.00011），证明该范式在梵语中具备有效性和可靠性。研究进一步揭示了语言特性对模型配置的影响，并为未来改进提供了方向。
链接: https://arxiv.org/abs/2609.25012
作者: Tanay Agrawal
机构: The Harker School (哈克学校)
类目: Computation and Language (cs.CL)
备注:
Abstract:Diachronic word embeddings have become the modern standard for tracking semantic change, yet they have been largely validated on modern, high-resource, and well-segmented languages. This paper tests whether the paradigm transfers to Sanskrit, an ancient, low-resource language whose phonological fusion (sandhi), morphological inflection, compounding, and polysemy pose a unique challenge. I assemble a 2.7M-token corpus spanning four canonical periods, recover word boundaries with a neural byte-level sandhi splitter and lemmatizer, and train per-period embeddings across configurations. To evaluate the system, I curate a validation set from historical scholarship and test recovery directionally with anchor displacement. Of 21 testable shifts, 19 move in the philologically attested direction (sign test, p=0.00011). I further show which configuration the language forces and comment on opportunities for improvement.
[NLP-90] Do Synthetic Personas Predict Real Audience Response? A Sim-to-Real Study Where a No-Persona Baseline Beats Persona-Based Copy Simulation
【速读】： 该论文旨在解决生成式 AI 在营销场景中作为“合成人格”（synthetic personas）用于预测受众对文案反应的可信度问题，核心关切在于：基于人物设定的 LLM 预测是否能有效映射真实用户行为。其解决方案的关键在于通过大规模真实 A/B 测试数据（来自 Upworthy Research Archive）构建“从模拟到现实”的有效性验证框架，对比了基于真实人口统计特征构建的十人设面板与无角色设定的零样本基线（zero-shot baseline）在预测点击率上的表现。研究发现，真实行为数据中仅有 399 组具有统计显著性的结果可作为有效评估基准；更关键的是，与预期相反，人物设定条件反而降低了预测有效性——零样本基线在 Kendall τ 相关性（0.361，中等效应）和首选准确率（49.2%）上显著优于人设面板（τ = 0.084，准确率 34.6%），且置信区间不重叠。这表明强制模型代入特定人格会引入偏差与噪声，而直接询问模型对典型读者的点击可能性，反而能更准确捕捉群体层面的先验分布。该结论在多个数据划分、不同领域数据集及多种模型（Gemini 多层级与 OpenAI gpt-4.1）下均保持一致，证实了“单纯使用大语言模型进行排序”优于依赖合成人格的模拟策略，揭示出合成人格不仅为弱预测工具，甚至可能劣于不使用它们。
链接: https://arxiv.org/abs/2609.25010
作者: Alexandre Cristovão Maiorano
机构: Lumytics(卢米提克斯)
类目: Artificial Intelligence (cs.AI); Computation and Language (cs.CL); Computers and Society (cs.CY)
备注: 12 pages, 3 figures, 7 tables
Abstract:Marketers increasingly use large language models (LLMs) as “synthetic personas” to predict how an audience will react to a piece of copy before it ships, encouraged by evidence that profile-conditioned LLMs mimic human samples. But is that prediction actually valid against real behaviour - and does the persona machinery help? We present a sim-to-real validity study using the Upworthy Research Archive - thousands of headline A/B tests on shared real traffic, with measured click-through - as held-out ground truth. We compare a ten-persona panel, grounded in the real audience’s demographics, against a no-persona zero-shot baseline that simply asks the model how likely a typical reader is to click. Two findings stand out. First, ground-truth reliability is the binding constraint: most A/B tests have no statistically distinguishable winner, so validity can only be measured on the reliable subset (n = 399). Second, and counter to the persona-simulation premise, persona conditioning degrades predictive validity: the no-persona baseline ranks variants markedly better (Kendall \tau = 0.361, a medium effect; top-1 accuracy 49.2%) than the persona panel (\tau = 0.084; top-1 34.6%), with non-overlapping confidence intervals. Asking the model directly taps an accurate population-level prior; forcing it to role-play specific personas injects bias and noise. The result replicates across three independent Upworthy splits, holds in direction on a different-domain news dataset, and is robust to seed, prompt phrasing, and model choice - across three Gemini tiers and a different model family (OpenAI gpt-4.1, significant paired gap). The takeaway: for predicting aggregate engagement, a plain LLM ranker beats persona simulation - synthetic personas are not merely a weak predictor, they are worse than not using them. All numbers regenerate from a public, artifact-first replication package.
[NLP-91] Same Quantity Different Answer: Numerical Representation Invariance in Language Models
【速读】： 该论文旨在解决生成式人工智能（Generative AI）在处理数值等价问题时的语义一致性与形式不变性问题，即同一数量以不同表达形式（如小数、分数、百分比、数字词、科学记数法或单位换算）呈现时，模型应保持输出结果的一致性。其核心挑战在于评估模型在多种语法和语义变换下的“轨道不变性”（orbit invariance），即对等价输入产生一致正确答案的能力。解决方案的关键在于构建一个包含3,600个精确有理数问题和8,600个跨五类保恒变换的提示集，并通过严格的语法审计（syntax audit）统一答案格式，避免依赖大语言模型（LLM）判断。实验发现，尽管规范答案的准确性（canonical accuracy）较高（0.969–0.996），但轨道正确率（orbit correctness）和轨道不变性（orbit invariance）显著下降至0.848–0.981，暴露出模型在科学记数法乘法形式上的语法支持缺失问题，导致看似推理失败实为接口设计缺陷。此外，研究揭示了另一种语义病理：Mistral Small 4在单位换算输入下表现骤降，准确率为0.699，并出现265例因数量级错误而偏离标准答案的系统性偏差。进一步实验表明，在低误差子集上，“表示共识”（representation consensus）并未优于“改写共识”（paraphrase consensus），且误报率更高。研究强调，评估框架的设计必须区分真正的推理缺陷与由表达形式不兼容引发的虚假失败。
链接: https://arxiv.org/abs/2609.25009
作者: Ephraim Atta-Duncan
机构: 未知
类目: Computation and Language (cs.CL); Machine Learning (cs.LG)
备注: 15 pages, 2 figures. Ancillary archive includes the frozen benchmark, evaluation and audit records, consensus raw responses, manifests, analysis code, and tests
Abstract:Numerically equivalent word problems should yield the same canonical answer whether a quantity is written as a decimal, fraction, percentage, number word, scientific notation, or an exactly converted unit. We generate 3,600 exact-rational problems and 8,600 prompts spanning five identity-preserving transformation families, and evaluate five open-weight systems. After a fixed syntax audit that normalizes common answer forms without an LLM judge, canonical accuracy is 0.969-0.996, but orbit correctness falls to 0.848-0.981 and orbit invariance to 0.851-0.981; invariant-but-wrong orbits account for at most 0.003. Most of the broad strict-parser collapse arises because multiplication-form scientific notation lies outside the implemented number grammar, illustrating how evaluator interfaces can masquerade as reasoning failures. A distinct semantic pathology remains: Mistral Small 4 scores 0.699 on unit-converted inputs and produces 265 errors differing from the label by exact powers of ten. In a separate 9,000-call experiment that allocates equal calls to the compared arms, representation consensus does not outperform paraphrase consensus on a low-error subset and produces substantially more false alarms. The accompanying ancillary archive contains the frozen benchmark, evaluation and audit records, consensus raw responses, manifests, analysis code, and a one-command paper build.
[NLP-92] raining a Language Model End-to-End in Rust: An Experience Report
【速读】： 该论文旨在解决在纯 Rust 语言环境中实现端到端大语言模型（Language Model, LM）预训练的可行性问题，尤其关注在无 Python、无 PyTorch 支持且由单人独立完成的情况下，利用租用 GPU 资源进行大规模训练所面临的系统性挑战。其核心问题是当前主流 Rust 机器学习框架（Candle 与 Burn）在训练后端支持上的严重缺陷，这些缺陷表现为隐蔽性故障（silent failures），如融合核函数（fused kernels）完全不产生梯度、反向传播吞吐量仅为理论值的约 3%、以及在百亿参数级别训练过程中因内核融合路径崩溃（segfault）等问题，且均未被常规损失曲线检验所揭示。解决方案的关键在于提出一种可推广的验证规范——“梯度流仲裁器”（gradient-flow arbiter），即通过执行一次前向-反向传播并强制断言所有可训练参数均获得有限且非零梯度，从而有效捕获此类隐性错误。此外，研究还揭示了孟加拉语脚本下分词器“生育陷阱”（tokenizer-fertility trap）导致的语料语言平衡失衡问题，并通过改进分词策略修复。最终训练出的约 0.4B 参数模型在孟加拉语建模任务上表现优异（每标记负对数似然为 0.93），但在英语常识选择题上仅达到随机水平，符合预期设计目标。研究表明，尽管 Rust 在模型推理部署方面具备潜力，但现阶段尚不具备与 PyTorch 等成熟框架竞争的训练能力。
链接: https://arxiv.org/abs/2609.25008
作者: Arif Adito
机构: Adioris Tech Ltd. (Adioris科技有限公司)
类目: Computation and Language (cs.CL)
备注:
Abstract:I pretrained a language model end-to-end in Rust - alone, with no team, no PyTorch, and no Python in the training path - for 164 in rented GPU time. I report that as an achievement, not a recommendation: the more useful contribution is a measured failure taxonomy of the two leading Rust ML frameworks, Candle and Burn, as training (not inference) backends in 2026. I document five Candle defects, including fused kernels that silently produce no gradient, and three Burn defects, including a backward pass at roughly 3% of theoretical GPU throughput and a kernel-fusion path that segfaults mid-training at multi-billion-parameter scale. Every one passed ordinary loss-curve inspection; none announced itself. I describe the verification discipline that caught six such silent failures, centered on a gradient-flow arbiter: a test that runs one forward/backward pass and asserts every trainable parameter receives a finite, nonzero gradient, generalizable to any framework. The trained model (roughly 0.4B parameters, Bangla-first) shows strong Bangla language-modeling signal - a per-token negative log-likelihood of 0.93 against 12.60 for a random-initialized twin - while scoring at chance on English commonsense multiple-choice, the expected outcome of a deliberately small, Bangla-weighted budget (about 2 billion tokens, 54.6 hours, one rented H100). I also report a tokenizer-fertility trap in Bengali script: naive byte-level tokenization collapsed Bangla to roughly 1.4 characters per token against English’s 3.9, silently inverting the corpus’s language balance; fixing it reached roughly 4.1. To my knowledge, this is among the first documented end-to-end LM pretraining runs in pure Rust. After this run I moved training to PyTorch and kept Rust for on-device serving: in my hands, Rust is not yet a competitive place to train a language model, though it may be a good place to serve one.
[NLP-93] What Does 99% Accuracy Measure? A Reproducible Audit of Shortcut Learning in a Widely Used Fake News Corpus
【速读】： 该论文旨在解决当前文本分类任务中，基于ISOT/Kaggle“假新闻与真新闻”数据集训练的分类器报告的高准确率（F1 > 0.98）与其实际评估信息真实性能力之间存在的矛盾问题。其核心问题是：这些看似优异的性能指标可能并非源于对内容真实性的有效判别，而是由数据集中存在的多种泄露通道（leakage channels）和分布偏移（distribution shift）所导致的虚假信号。解决方案的关键在于通过透明的TF-IDF与线性分类器管道对数据集进行系统性审计，识别并量化三类主要泄露源——标题元数据、新闻社标签（新传社标签在真实文章中占比99.2%）、以及6,251个重复文档（污染了19.4%的原始测试集）。研究发现，仅使用标题元数据即可实现F1 = 1.000，表明类别间存在完全分离的标题主题；即使清除所有泄露通道后，模型仍保持高达0.9814的F1，说明剩余信号主要来自模糊的编辑风格而非明确的关键词；更重要的是，这种风格信号不具备跨主题迁移能力，在主题不重叠的测试协议下，平均精度下降5.2点，部署F1降至0.8067，且在独立的LIAR基准上所有模型均退化至近随机水平（ROC-AUC ≈ 0.55），无法超越多数类基线。因此，作者指出当前数据集上的高分主要反映的是来源与主题的可区分性，而非真实性的判别能力，并强调应采用元数据仅用、小样本及主题不重叠等低成本基线作为未来研究的诊断工具，以避免模型依赖于捷径学习（shortcut learning）。
链接: https://arxiv.org/abs/2609.25006
作者: Yuvraj Verma
机构: 未知
类目: Computation and Language (cs.CL); Machine Learning (cs.LG)
备注: 17 pages, 11 figures, 11 tables. Code, experiment scripts, and machine-readable results: this https URL
Abstract:Text classifiers trained on the ISOT/Kaggle “Fake and Real News” corpus routinely report accuracy and F1 above 0.98, a level of performance that sits uneasily beside the difficulty of assessing veracity. Using a transparent TF-IDF and linear-classifier pipeline as a measurement instrument, we audit the corpus along three leakage channels and two distribution-shift protocols, releasing all code and derived numbers. First, the benchmark is partly degenerate: a classifier given only the subject metadata field, with the article text discarded, attains F1 = 1.000, since the two classes have disjoint subjects. Second, removing all three leakage channels, metadata, a newswire source tag present in 99.2% of real articles, and 6,251 duplicate documents contaminating 19.4% of a naive test split, lowers F1 by only 1.21 points (0.9935 to 0.9814); the residual signal is diffuse editorial style rather than a few giveaway tokens, since deleting the 1,000 highest-weight unigrams still leaves F1 = 0.926. Third, this style signal does not transfer: under a topic-disjoint protocol, average precision falls from 0.9995 to 0.9475 and deployed F1 from 0.9905 to 0.8067, with a prior-matched analysis confirming a genuine 5.2-point loss of discrimination, while temporal transfer is nearly lossless. A fine-tuned DistilBERT is stronger in-distribution (F1 = 0.9993) but degrades far more under topic shift, losing 12.9 average-precision points against the linear model’s 5.2. Transferred to the independent LIAR benchmark, all three models fall to near-chance ranking (ROC-AUC 0.54-0.57), none beating a majority-class baseline. We conclude that within-corpus scores here quantify source and topic separability rather than veracity, that added capacity exploits the shortcut rather than avoiding it, and we recommend metadata-only, small-sample, and topic-disjoint baselines as inexpensive diagnostics for future work.
[NLP-94] When Residualization Helps an Audit: Format Effects Slice Gains and Their Limits
【速读】： 该论文旨在解决当前大语言模型（LLM）评估体系中评价分数易受表面形式（surface form）干扰而无法真实反映其本应衡量的内在质量（如代码正确性、推理合理性等）的问题。其核心挑战在于，现有评估机制（如奖励模型、重排序器和人工评分模型）常因对代码注释、格式或冗余文本等表面特征的过度敏感，导致评估结果偏离实际能力表现。论文提出的关键解决方案是采用残差化（residualization）方法以剔除可预测的表面成分，但指出仅依赖残差化不足以确保评估的有效性：被移除的表面信号可能包含与构念（construct）相关的有效信息，且残差化无法区分哪些成分属于构念相关、哪些仅为表面噪声。通过在单元测试标签与仅修改注释的干预实验中验证，残差化虽能减弱奖励模型对格式的偏好约0.12，但对正确与错误代码之间的判别力提升不足0.01，表明其效果有限。在自然语言推理（NLI）与问答（QA）的观测场景中，使用独立标注者标签重新评估冻结的样本，发现尽管在特定子集上与构念标签的一致性有所提升，但整体人群的一致性普遍下降，且题内排序性能亦未改善。研究进一步揭示，当构念特征与表面特征高度纠缠时，残差化可能在降低表面相关性的同时损害构念对齐性，甚至存在若干配置在调整前所有检验中均通过，却严重削弱构念有效性的情况，说明缺乏“保证性”校验机制。因此，论文最终提出一种报告协议：任何经过调整的评分结果必须伴随审计时刻的诊断说明，明确指出其构念对齐性的代价，而非作为原始分数的替代品，强调透明性与批判性评估的重要性。
链接: https://arxiv.org/abs/2609.24194
作者: Daein Weon,Dong Ho Kang
机构: Kookmin University (国立酷比大学); UStechlab (UStech实验室)
类目: Computation and Language (cs.CL); Machine Learning (cs.LG)
备注: 61 pages, 4 figures, 40 tables. Code: this https URL
Abstract:Evaluation scores used around LLM systems – including reward models, rerankers, and LLM judges – can track surface form instead of the quality they claim to measure. When presented with a terse correct solution and a commented buggy solution for the same MBPP problem, a public preference reward model selects the correct one no better than a coin flip (0.507). Subtracting the predictable surface component from such scores is increasingly common, but removal alone does not yield a more valid measurement: the removed component may carry construct-relevant signal, and residualization cannot tell which is which. Under designed interventions – unit-test labels with comment-only edits – residualization attenuates the reward model’s format effects by about 0.12 on both correct and buggy code, while the correct-versus-buggy margins move by less than 0.01. In observational NLI and QA settings, we freeze a held-out replication before scoring and re-evaluate it using labels from disjoint annotators; this supports only a narrower conclusion: better agreement with the construct labels on a pre-declared slice where a surface-only predictor errs, not a repaired score. Full-population agreement falls in every observational setting with a reported positive slice gain, and within-question ranking falls in every such QA setting. When construct and surface features are entangled, residualization can decorrelate a score while degrading construct alignment, and, in a controlled model, configurations just as damaging to construct alignment pass every pre-adjustment check, so no committed gate is a guarantee. We assemble these distinctions into a reporting protocol whose outcomes, refusal included, state what an adjusted score may be claimed to show: an audit-time diagnostic reported beside the construct-alignment cost it incurs, never a replacement for the raw score.
[NLP-95] Are Human-Aligned Models Models of Humans? A Turing-Test Gap in Preference Alignment
【速读】： 该论文旨在解决当前生成式人工智能（Generative AI）在对齐人类反馈时存在的根本性问题：即模型对齐人类偏好（human preferences）并不必然导致其行为具有人类相似性（human-likeness）。研究指出，人类所偏好的回答未必是他们自身会给出的回答，因此将模型对齐于人类偏好与对齐于人类实际行为之间存在本质差异。作者提出“图灵测试差距”（Turing-test gap）这一概念，用以描述偏好对齐可能导致模型输出偏离真实人类响应分布的现象。其解决方案的关键在于揭示：只有在特定严格条件下，偏好对齐才能保持人类响应分布的保真度；而实证分析表明，现实中的真实人类偏好通常不满足该条件，且随着偏好权重增强，无论方向如何，模型生成内容的人类似度均显著下降。此外，标准直接偏好优化（DPO）方法同样表现出该差距。研究结果明确将“人类相似性”确立为对齐维度中的一个独立、需显式考虑的目标，而非默认由偏好对齐自然实现的副产品。
链接: https://arxiv.org/abs/2609.23640
作者: Suqin Yuan,Runqi Lin,Muyang Li,Guanzhe Hong,Jindong Gu,Lei Feng,Chris Russell,Tongliang Liu
机构: University of Oxford(牛津大学); Southeast University(东南大学); University of Sydney(悉尼大学)
类目: Artificial Intelligence (cs.AI); Computation and Language (cs.CL); Machine Learning (cs.LG)
备注:
Abstract:Human-feedback alignment has made language models useful assistants and is commonly described as aligning them with humans. However, the responses people prefer from an AI need not be the responses they themselves would give. We distinguish alignment with human preferences from alignment with human behavior, and show that alignment with human preferences can make model behavior less human-like even when both preferences and responses come entirely from humans. We call this the Turing-test gap. We show that preference alignment preserves the human response distribution only under a restrictive condition, and find no consistent evidence that real human preferences satisfy it. Empirically, the loss of human-response likelihood increases with the strength of preference weighting, regardless of its direction, and the gap also appears under standard DPO. These results establish human-likeness as an explicit dimension of alignment rather than something assumed to follow from preference alignment.
[NLP-96] Advanced spectral clustering for heterogeneous data in credit risk monitoring systems
【速读】： 该论文旨在解决异质性数据（heterogeneous data）在信用监控中带来的挑战，即如何有效融合数值型金融变量与文本记录两类异构信息以实现精准的信用风险分群。其核心解决方案是提出一种先进的谱聚类方法——高级谱聚类（Advanced Spectral Clustering, ASC），该方法通过优化权重参数协同整合金融与文本相似性，并采用基于特征值-轮廓系数（eigenvalue-silhouette）的新型特征向量选择策略，显著提升了聚类质量。实验结果表明，ASC在包含1,428家中小企业的数据集上相较于单一类型数据基线方法，轮廓系数（Silhouette score）提升18%，且在多种聚类算法（如k-means、k-medians、k-medoids）下均表现出高度稳定性（ΔIntra/Inter < 0.13，ΔSilhouette Coefficient < 0.02）。此外，聚类结果揭示了具有实际业务意义的模式，例如“社会招聘”关键词在低风险企业文本记录中出现频率达51%，并识别出以招聘为核心特征的中小企业群体其违约风险降低30%。因此，ASC的关键在于通过理论与异质数据应用的深度融合，实现对多源信息的有效融合与可解释性聚类，从而支持更精准、可操作的信贷干预决策。
链接: https://arxiv.org/abs/2509.00546
作者: Lu Han,Mengyan Li,Jiping Qiang,Zhi Su
机构: 未知
类目: Machine Learning (cs.LG); Computational Engineering, Finance, and Science (cs.CE); Computation and Language (cs.CL)
备注: 25 pages, 7 figures, 6 tables
Abstract:Heterogeneous data, which encompass both numerical financial variables and textual records, present substantial challenges for credit monitoring. To address this issue, we propose Advanced Spectral Clustering (ASC), a method that integrates financial and textual similarities through an optimized weight parameter and selects eigenvectors using a novel eigenvalue-silhouette optimization approach. Evaluated on a dataset comprising 1,428 small and medium-sized enterprises (SMEs), ASC achieves a Silhouette score that is 18% higher than that of a single-type data baseline method. Furthermore, the resulting clusters offer actionable insights; for instance, 51% of low-risk firms are found to include the term ‘social recruitment’ in their textual records. The robustness of ASC is confirmed across multiple clustering algorithms, including k-means, k-medians, and k-medoids, with \DeltaIntra/Inter 0.13 and \DeltaSilhouette Coefficient 0.02. By bridging spectral clustering theory with heterogeneous data applications, ASC enables the identification of meaningful clusters, such as recruitment-focused SMEs exhibiting a 30% lower default risk, thereby supporting more targeted and effective credit interventions.
[NLP-97] Qwen -Audio-3.1-Realtime: Towards Reliable Agent ic Voice Interaction
【速读】： 该论文旨在解决实时语音助手在动态对话过程中需同时实现多轮推理、动作执行与对话规则遵循的复杂挑战，尤其关注语音理解、多语言处理、工具调用、对话行为一致性及全双工交互中的抗干扰能力。其核心解决方案在于提出“思考（Think）、行动（Act）、说话与协调（Speak and Coordinate）”三阶段协同架构：关键创新在于通过核心混合监督微调（Core-Cocktail SFT）结合多模态与多教师在线策略蒸馏（M²-OPD），实现语言能力迁移并原生构建音频感知技能；在行动阶段，采用自演化可执行环境与多粒度回溯机制，基于**组相对策略优化（GRPO）**提升模型对工具的使用能力、反馈解析与任务闭环完成能力；在说话与协调层面，系统性建模了语音响应的时机、方式与触发条件，以增强交互自然性与鲁棒性。实验表明，相较于Qwen-Audio-3.0-Realtime，Qwen-Audio-3.1-Realtime在半双工语音转文本任务上的整体任务成功率由78.4%提升至82.0%，而在全双工语音交互基准中，对背景语音的误响应率从73.0%显著下降至13.0%。此外，研究还提出了独立的语音控制原型（Voice Harness），利用前景-背景协同与记忆机制，将语音交互扩展至持续性任务场景，进一步验证了该框架在复杂现实应用中的可行性与优越性。
链接: https://arxiv.org/abs/2609.25176
作者: Lujia Bao,Qian Chen,Luyao Cheng,Chong Deng,Yuxiang Kong,Xiangang Li,Xu Li,Jiaqing Liu,Chao-Hong Tan,Haoyu Wang,Wen Wang,Xilou Wang,Junhao Xu,Liang Yi,Binbin Zhang,Qinglin Zhang,Qiquan Zhang
机构: Alibaba Token Foundry, Alibaba Group(阿里巴巴集团)
类目: Audio and Speech Processing (eess.AS); Artificial Intelligence (cs.AI); Computation and Language (cs.CL); Sound (cs.SD)
备注: 20 pages, technical report
Abstract:Real-time voice assistants must reason over evolving requests, execute actions, and follow conversational rules. Qwen-Audio-3.1-Realtime brings these requirements together through Think, Act, and Speak and Coordinate. Think combines Core-Cocktail supervised fine-tuning with Multimodality and Multi-Teacher On-Policy Distillation (M ^2 -OPD) to transfer language capabilities and develop native audio skills. Act uses self-evolving executable environments and multi-granularity rollouts for Group Relative Policy Optimization (GRPO), teaching the model to use tools, interpret feedback, and complete tasks. Speak and Coordinate aligns how, when, and whether the assistant speaks or acts. We evaluate audio reasoning, multilingual understanding, tool use, conversational behavior, full-duplex interaction, and safety. Compared with Qwen-Audio-3.0-Realtime, 3.1 raises overall task success from 78.4% to 82.0% on our half-duplex speech-to-text adaptation of \tau -Voice. On speech-to-speech Full-Duplex-Bench v1.5, the response rate to background speech falls from 73.0% to 13.0%. We also present a separate Voice Harness prototype, using Qwen-Audio-3.0-Realtime as its foreground, that extends spoken interaction to persistent tasks through foreground–background coordination and memory.
[NLP-98] Beyond Short Segments : Expanding Speaker Embeddings with Vector Archives INTERSPEECH2026
【速读】： 该论文旨在解决当前先进说话人验证（Speaker Verification, SV）系统在短时语音段上性能显著下降的问题，其根本原因在于短语音中缺乏足够的说话人特异性信息。为应对这一挑战，提出了一种名为向量档案映射ECAPA（Vector Archive Mapping ECAPA, VAM-ECAPA）的新架构，其核心创新在于引入基于Transformer的向量档案映射与统计池化模块（Transformer-based Vector Archive Mapping with Statistical Pooling, TVAMSP）。该模块通过将低信息量的短语音特征映射到一个可学习的“向量档案”（Vector Archive），该档案存储了典型说话人的标准化声学表征，从而有效增强原始特征的判别能力。通过将TVAMSP模块嵌入强大的WavLM+ECAPA-TDNN基线模型，系统能够从短时语音中生成鲁棒且具有高区分性的说话人表示。在VoxCeleb1基准测试中，VAM-ECAPA在1秒测试片段上取得了8.334%的等错误率（EER），相比传统训练基线实现了54.8%的相对错误率降低，验证了该方案在短语音场景下的有效性。
链接: https://arxiv.org/abs/2609.25007
作者: Hyunku Kang,Minkyu Cho,Chanwoo Kim
机构: Korea University (韩国大学)
类目: Audio and Speech Processing (eess.AS); Computation and Language (cs.CL); Sound (cs.SD)
备注: Accepted at INTERSPEECH 2026 (oral)
Abstract:The performance of state-of-the-art speaker verification (SV) systems severely degrades on short utterances due to insufficient speaker-specific information. To address this critical challenge, we propose the Vector Archive Mapping ECAPA (VAM-ECAPA), a novel system designed to enhance feature extraction from short-duration speech. The core of our system is the Transformer-based Vector Archive Mapping with Statistical Pooling (TVAMSP) module, which enriches information-scarce features by mapping them against a learnable Vector Archive of canonical speaker traits. By integrating the TVAMSP module into a strong WavLM+ECAPA-TDNN baseline, our system learns to map sparse features from short segments into robust, discriminative speaker representations. Experiments on the VoxCeleb1 benchmark show that our proposed VAM-ECAPA achieves a highly competitive EER of 8.334% on 1-second test segments, a 54.8% relative error reduction compared to a conventionally-trained baseline.
信息检索
[IR-0] SpeakerMem-R1: Speaker-Centered Dual-Track Memory for Multi-Party Dialogue
链接: https://arxiv.org/abs/2609.26780
作者: Haobo Zheng,Tan Tang,Yan Chen,Weijie Wang,Yingcai Wu
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI); Information Retrieval (cs.IR); Machine Learning (cs.LG)
备注: Project Page: this https URL , Code: this https URL
Abstract:Long-term conversational memory in multi-party settings requires more than retrieving relevant content from long-term conversations: it must distinguish who said what, whom each statement concerns, how individuals perceive one another, what information is shared by the group, and how states change over time. Recent studies on multi-party dialogue benchmarks show that existing general-purpose LLM memory systems tend to lose person and group relations or struggle to integrate clues distributed across members, groups, and time. Together, these issues reveal two core bottlenecks: message attribution and relational understanding in multi-party dialogue, and state reconstruction from interleaved histories. To address both, we propose \textbfSpeakerMem-R1 : its dual-track memory stores speaker-labeled verbatim messages and derived states organized into person-level and group-level views, then combines evidence from both tracks by entity, event, and time at query time. To reduce attribution and update errors during structured memory construction while enabling local deployment, we train Writer-R1 with SpeakerLevenshtein and speaker-conditioned GRPO. On GroupMemBench, SocialMemBench, and EverMemBench, SpeakerMem-R1 achieves binary accuracies of 47.9%, 69.2%, and 61.9%, respectively. On the publicly reported EverMemBench leaderboard from EverMind-AI, we achieves 62.33%, the best reported result among the latest state-of-the-art frameworks. It also achieves 70.85% on all 1,986 LoCoMo questions, which we use as a two-person long-term conversation boundary test. In a controlled evaluation of 305 questions, RL raises the SFT Writer’s mean accuracy from 57.38% to 68.20%. We report both binary accuracy and token-F1, and ablations show that the verbatim and structured tracks, as well as person-level and group-level views, are complementary under the standardized evaluation interface.
[IR-1] Discovery-Driven Integration of Disjoint Tables via Text
链接: https://arxiv.org/abs/2609.26658
作者: Md Ataur Rahman,Dimitris Sacharidis,Oscar Romero,Sergi Nadal
类目: Information Retrieval (cs.IR); Computation and Language (cs.CL); Databases (cs.DB); Machine Learning (cs.LG)
备注:
Abstract:Integrating heterogeneous datasets within data lakes is a critical challenge, particularly for semantically related tables that lack the explicit attributes needed to be joined. We study Discovery-Driven Integration, where the relevant sources and their missing relational structure must be discovered before integration. In this setting, unstructured text provides the evidence that connects otherwise disjoint tables. The fundamental challenge is to discover the relationships at a fine-grained level that connect individual rows from different tables through specific sentences. We formalize this task as Text-Mediated Join Path Discovery and propose a horizontal bidirectional cross-attention architecture called LOKI Latent-space Optimization for Knowledge Integration) that learns contextualized representations of table rows and sentences. Through a global table-text contrastive objective, fine-grained row-sentence associations emerge without explicit local supervision. Existing multi-modal discovery methods largely retrieve coarse-grained column-text associations, whereas integration systems assume supplied row-text links, schemas, or queries. LOKI instead transforms these implicit associations into explicit, interpretable join paths, organizes them into relation-consistent groups, and materializes them as typed integrated tables with sentence-level provenance. Comprehensive evaluations on real-world benchmarks demonstrate that LOKI consistently outperforms state-of-the-art multi-modal data discovery approaches, and materializes typed integrated tables with 0.982 macro typed-pair precision while being up to 40 times cheaper in LLM API cost than direct prompting.
[IR-2] When Does Permutation Instability Generalize? Independent-View Validation for Listwise LLM Reranking
链接: https://arxiv.org/abs/2609.26251
作者: Wenzhang Du(Independent Researcher)
类目: Information Retrieval (cs.IR)
备注: 8 pages, 2 figures, 4 tables
Abstract:Listwise language-model rerankers often disagree across equivalent candidate permutations. Finite instability diagnostics are therefore used to motivate additional sampling, aggregation, or selective computation. But an association with a validation statistic that reuses the probe views need not isolate predictive information about unseen permutations. Shared measurements can induce classical part-whole association. We study how this affects claims that a finite-view instability score predicts unseen permutations. We derive the exact finite-view decomposition and prospectively compare zero, one, and two reused views, including a fully disjoint four-view target. The study covers two pinned 7B model families and two recommendation datasets, with controlled lists for signed offline analysis and untouched retriever lists for target-free replication. On the four controlled blocks, fully disjoint correlations are weak or heterogeneous (-0.061 to 0.281), whereas reusing both probe views yields 0.600 to 0.718; all paired contrasts are large (0.436 to 0.661) and Holm-significant. The overlap effect is positive in all four untouched-list blocks. Increasing the probe from two to four views clearly improves disjoint reliability in only one block. Moreover, the probe predicts aggregation-movement magnitude (Spearman rho = 0.142 to 0.426) but not stable signed target benefit, and 7 of 12 fixed-fraction probe-routing points are strictly dominated at measured cost. Thus, when the intended estimand is predictive information about unseen perturbation behavior, validation targets must be observation-disjoint from the probe to isolate that information; signed utility and cost-sensitive decisions remain separate questions. Comments: 8 pages, 2 figures, 4 tables Subjects: Information Retrieval (cs.IR) Cite as: arXiv:2609.26251 [cs.IR] (or arXiv:2609.26251v1 [cs.IR] for this version) https://doi.org/10.48550/arXiv.2609.26251 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
[IR-3] Which Reranking Conclusions Survive the Answer Interface? A Prospective Finite-Orbit Audit
链接: https://arxiv.org/abs/2609.26250
作者: Wenzhang Du
类目: Information Retrieval (cs.IR)
备注: 7 pages, 1 figure, 4 tables
Abstract:Rerankers are increasingly evaluated through downstream language-model answers. This raises a retrieval-measurement question: if only the reader’s answer interface changes, should we reach the same conclusion about BM25 versus BGE-v2-m3? We prospectively audit their claim-paired effect on RAGuard and FEVER with four readers. Retrieval policies, evidence, claims, and context depth remain fixed while semantic-to-label binding, A/B versus X/Y vocabulary, and option order form eight task-equivalent interfaces. We ask whether the estimated retrieval-policy effect, its ordering, or selection value changes. None of the six confirmatory settings showed statistically certified interface variation above the prespecified 0.015 materiality threshold, and none showed a certified reversal of the BM25-BGE ordering. Selector disagreement reaches 33.5% in one environment, yet none of eight environments establishes the prespecified material held-out value difference. These results do not support broad replicated instability, but they do not prove universal invariance: five settings remain too uncertain to satisfy the prespecified higher-order equivalence condition. They show why retrieval evaluations should separate policy level, interface stability, policy ordering, and selection value. When stability is unverified, a uniform average over the enumerated interfaces with explicit variation bounds avoids privileging one interface.
[IR-4] ABAI at COLIEE 2026 Task 1: Multi-Stage Retrieval with GraphRAG -Enhanced Meta-Learning and a Post-Hoc Study of the Cross-Validation-to-Test Gap
链接: https://arxiv.org/abs/2609.26237
作者: Minhan Cho,Soyoung Park,Daejin Choi,Jinyoung Han
类目: Information Retrieval (cs.IR); Artificial Intelligence (cs.AI); Computation and Language (cs.CL)
备注: 13 pages, 3 figures, 9 tables. Extended version of the paper presented at COLIEE 2026 (Workshop on the Thirteenth International Competition on Legal Information Extraction and Entailment), Singapore, June 2026. Code: this https URL
Abstract:We present the ABAI submission to COLIEE 2026 Task 1, case law retrieval, together with a controlled study of why it underperformed. The task suppresses the cited passages themselves, which removes much of the lexical overlap a retriever would rely on. Our pipeline answers this with four independently trained stages: multi-view BM25 over citation-context windows with reciprocal rank fusion, neural reranking, graph-based features from entity communities and a graph attention network, and a LightGBM meta-learner over 34 features. Our best run reached F1=0.177 on the official test set, against a cross-validated 0.311, and we attributed that gap to a recall ceiling, temporal distribution shift, and threshold miscalibration. We then tested all three. Under leakage-free protocols threshold transfer costs 0.007 F1, decision quality is flat across chronological quartiles, and the official test queries are not measurably farther from the training manifold than training queries are from each other, in two independent embedding spaces. Decomposing the misses instead splits them exactly evenly between candidates never retrieved and candidates retrieved but ranked below the cut. Measuring the remedies for each half, BM25 length-normalisation tuning, an event-triple view, and full-content dense fusion lift top-200 recall by three to seven points, and citation-graph features add 0.014 F1 over eight seeds once own-citation leakage is removed, while per-query cutoff rules, a zero-shot reranker swap, and a date filter do not help. We also document four evaluation artifacts, each of which reversed a result once the protocol was corrected.
[IR-5] A Semantic Approach to the Academic Publishing Network: Document Vector Representations and Hybrid Structural-Semantic Fusion over OpenAlex Data
链接: https://arxiv.org/abs/2609.26218
作者: Robert Šamárek,Radek Martinek
类目: Information Retrieval (cs.IR); Computation and Language (cs.CL); Digital Libraries (cs.DL)
备注:
Abstract:Structural graph analysis of the academic publishing network captures the topological relationships between entities but does not see the content of works. Building on our structural approach, this work complements it with a semantic layer and a parameterized structural-semantic fusion. We represent scientific documents by citation-informed vector embeddings (SPECTER2) and store them in an embedded vector database keyed by the stable OpenAlex ID, so that they connect directly to the graph layer. We define a modular late-fusion function that combines semantic similarity (cosine of embeddings) and structural similarity (bibliographic coupling) with a tunable weight alpha whose value is chosen according to the specific task. On the corpus of VSB - Technical University of Ostrava we show two things: citation-informed embeddings agree with the expert OpenAlex topical taxonomy better than a TF-IDF baseline, and in a recommendation use case the structural, semantic, and combined signals carry information in different regimes depending on the available data. Hybrid fusion here is not a universally better method but an explicit mechanism for steering complementary signals according to the task. We release the whole approach as an open-source extension of the apnet library with a reproducible workflow.
[IR-6] When Concealed Links Cannot Be Recovered: A Structural Identifiability Bound and Evaluation Pitfalls in Offshore Leak Networks
链接: https://arxiv.org/abs/2609.26171
作者: Joseph Bingham
类目: Information Retrieval (cs.IR); Probability (math.PR)
备注: 9 pages, 7 figures
Abstract:Leaks such as the Panama and Paradise Papers expose large networks of offshore entities, and they invite an obvious question for network learning. Can the relations these structures are built to hide—above all, who beneficially owns what—be recovered from the public part of the leak by link prediction? We argue that the answer is mostly no, and that the analyses which suggest otherwise are measuring the wrong thing. Our main result is a distribution-free identifiability bound. For any recovery rule that respects graph isomorphism, and that covers every topological link-prediction score together with every message-passing graph neural network, a concealed endpoint left isolated in the observed graph is interchangeable with its structural twins, so its hidden edge cannot be recovered above chance. Isolation is only the sharpest case. In general the ceiling on recovery is set by the size of a node’s structural-indistinguishability class, for which the Weisfeiler–Leman colour class is a computable stand-in, and node degree is at best a loose proxy. On the full ICIJ Offshore Leaks graph (814K entities and 84K labelled beneficial-owner edges) a classifier-free, degree-controlled probe reproduces an exact 0.5 floor for isolated owners, who make up 20.6% of all owners, and a trained graph neural network lands on the same floor. Recovery climbs only as structural distinctiveness grows, and the floor reappears in every one of the five leaks. Along the way we document five evaluation traps. Each one makes a bound that cannot be beaten look beaten, and we give a short rule that avoids them. The practical upshot is to redirect effort from the hidden principal, which is close to unrecoverable, toward the machinery of concealment, and to spell out why fusing external data helps far less than one would hope.
[IR-7] ailSpec-EASE: Knowledge-Graph-Regularized Linear Recommendation for Web Long-Tail Discovery
链接: https://arxiv.org/abs/2609.26143
作者: Jianru Shen
类目: Information Retrieval (cs.IR); Machine Learning (cs.LG)
备注: Accepted at the main research track of WISE 2026 (26th International Conference on Web Information Systems Engineering)
Abstract:Recommender systems on Web platforms tend to over-serve popular items and neglect the long tail. Item-side knowledge graphs (KGs), often available as linked data or RDF-style Web resources, can help by connecting sparse items through shared semantic attributes. Many competitive KG-aware recommenders rely on graph neural architectures, whereas strong shallow linear models such as EASE-R typically ignore side information and may become infeasible in their global closed-form version. We introduce TailSpec-EASE, a lightweight recommender that injects a relation-aware spectral KG prior into a local closed-form reconstruction objective. The prior strength adapts to item popularity, giving stronger semantic guidance to long-tail items. Across four public benchmarks and a broad set of classical, linear, graph-CF, KG-aware neural, and score-level KG baselines, TailSpec-EASE attains a favorable trade-off between overall accuracy, long-tail performance, and training cost. It improves NDCG@20 by up to 24% over its counterpart without KG information. All tail-metric improvements over the no-KG counterpart are statistically significant under a paired bootstrap, and overall NDCG improves significantly on three of the four datasets. In a representative Amazon-book timing study, TailSpec-EASE trains in 37 seconds on CPU, compared with 2,584 seconds for a GPU-trained KGAT run and 15,800 seconds for CPU LightGCN, while attaining higher NDCG@20 and Tail Recall@20 on that dataset. It also remains feasible on catalogs where the global closed-form model runs out of memory.
[IR-8] CoVeR: Coverag e-Based Routing of Verifier Calls in Agent ic Retrieval
链接: https://arxiv.org/abs/2609.26086
作者: Daeyoung Roh,Donghee Han
类目: Computation and Language (cs.CL); Information Retrieval (cs.IR)
备注: 34 pages, 4 figures. Preprint. Under review
Abstract:An agentic retrieval system issues a sequence of search queries and must decide, at each step, whether the evidence collected so far is enough to stop. Delegating that decision to an LLM verifier or a prompt judge makes stopping reliable, but the verifier then reprocesses the growing evidence after every retrieval step, a substantial repeated cost. We show that most of these calls can be skipped without materially changing answer accuracy: a single threshold on a frozen sentence-embedding coverage margin detects the states in which the evidence is still plainly incomplete, and the verifier is called only on the ambiguous remainder, a gate we call CoVeR (Coverage-based Verifier Routing). Across three multi-hop QA benchmarks, with the evaluation protocol fixed before the full-scale run, the CoVeR-gated agent matches the answer accuracy of both the full-budget agent and the always-verify baseline within a fraction of an EM point. It cuts 62-68% of verifier calls, and 93% in a saturated regime. Routers built on evidence counts, lexical overlap, or BM25 relevance, alone or learned in combination, give weaker overall trade-offs, the gate transfers without re-tuning across deciders and agent scales, and its drafter distills into a 921k-parameter head atop the frozen encoder, leaving no LLM in the routing loop. The same signal cannot replace verification: matching a claim is far easier than deciding the claim is supported.
[IR-9] Knowledge-as-Skill: A Structural Design for Autonomous Knowledge-Base Use by LLM Agents
链接: https://arxiv.org/abs/2609.25991
作者: Jiangxu Wu
类目: Information Retrieval (cs.IR)
备注:
Abstract:Retrieval-augmented generation (RAG) gives large language models (LLMs) access to external knowledge, but its conventional retrieve-concatenate-generate pipeline makes retrieval decisions on behalf of the model. As tool use and agent loops become more reliable, an agent can decide whether to retrieve, what to inspect, and when to stop. This shift exposes a new bottleneck: the agent may not know what a knowledge base contains. Traditional knowledge bases expose documents as anonymous text chunks with limited information about scope, purpose, provenance, or relations. We propose Knowledge-as-Skill, an organization scheme that makes a knowledge base discoverable, navigable, and self-descriptive. It has three layers: a discovery layer centered on this http URL; a navigation layer with one this http URL per directory; and a knowledge layer containing documents with YAML frontmatter for topic, type, provenance, and lifecycle. The design follows the Open Knowledge Format (OKF) and the Skill protocol without modifying the agent framework. We also provide knowledge-as-skill, a pipeline for converting heterogeneous collections of PDFs, Word files, web exports, and notes into this structure. In a preliminary evaluation on the WixQA enterprise customer-support benchmark, our setup obtains 0.889 Factuality and 0.816 Context Recall, compared with reported Corpus2Skill values of 0.767 and 0.708. It obtains slightly lower Faithfulness, lower Context Precision, and more interaction turns. Because the models, prompts, and knowledge-package construction differ, these results are directional cross-work evidence rather than a controlled comparison. Subjects: Information Retrieval (cs.IR) Cite as: arXiv:2609.25991 [cs.IR] (or arXiv:2609.25991v1 [cs.IR] for this version) https://doi.org/10.48550/arXiv.2609.25991 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
[IR-10] ARAFA: An LLM -Generated Arabic Fact-Checking Dataset
链接: https://arxiv.org/abs/2609.25833
作者: Christophe Khalil,Shady Elbassuoni,Rida Assaf
类目: Computation and Language (cs.CL); Information Retrieval (cs.IR)
备注:
Abstract:Automatic fact-checking poses a significant challenge in Arabic natural language processing due to the scarcity of datasets and resources. In this manuscript, we introduce Arafa, a new large-scale dataset for fact-checking in Modern Standard Arabic, constructed through an automated framework leveraging large language models (LLMs). The dataset was constructed through a three-step pipeline: (1) claim generation from Arabic Wikipedia pages with supporting textual evidence, (2) claim mutation to generate challenging counterfactual claims with refuting evidence, and (3) an automatic validation step to validate that the generated claims are either supported or refuted by their accompanying evidence, or if the evidence does not provide enough information to judge the validity of the claims. The resulting dataset comprises 181,976 claim-evidence pairs labeled as supported, refuted, or not enough information. Human evaluation carried out on a test sample from the dataset demonstrated strong inter-annotator agreement (kappa = 0.89) using Cohen’s Kappa for supported claims and (kappa = 0.94) for refuted claims. Automatic validation based on a human-evaluated sample achieved 86% accuracy for supported claims and 88% for refuted ones. To showcase Arafa’s value as a resource for automatic Arabic fact-checking, four open-source transformer-based models were fine-tuned using Arafa, with the top-performing model achieving a Macro F1-score of 77% on the test data. In addition to Arafa being the first large-scale dataset for Arabic fact-checking, our framework presents a scalable approach for developing similar resources for other low-resource languages.
[IR-11] Robust Fusion of Semantic and Behavioural Signals for LLM Reranking in Personalised Search RECSYS2026
链接: https://arxiv.org/abs/2609.25825
作者: Aleksandr V. Petrov,Nathan Stein,Erik Lybecker,Emma Schüldt,Daniel Lazarovski,Hugues Bouchard,Mounia Lalmas
类目: Information Retrieval (cs.IR)
备注: Accepted at the USRW Workshop at RecSys 2026
Abstract:Personalised search must satisfy query intent while incorporating user context and historical interactions. LLM-based cross-encoders provide a single reranking interface, but injecting predictive behavioural statistics into their prompts can encourage shortcut learning: reliance on historical signals at the expense of semantic and user-context patterns that generalise to sparse or unseen searches. We study this problem in the personalised search system of a large-scale audio streaming platform using Query Slice Stats (QSS), an interaction-derived behavioural feature summarising historical success for query-candidate pairs. Naive QSS injection improves ranking when the feature is available but reduces robustness when it is removed. We address this with deterministic dual-sample feature-dropout training, which presents each example once with QSS included and once with QSS removed. Offline, QSS injection improves ranking quality by 13.3% when available. Dual-sample training preserves these gains while improving performance under QSS-removed evaluation by 4.0% relative to naive QSS training. In a live online test, both QSS-aware variants improve search success by roughly 2%. The aggregate test does not distinguish dual-sample from features-only training; the cold-start comparison is directionally consistent with the offline results. Paired feature-present and feature-removed training can therefore reduce the tension between exploiting strong behavioural statistics and remaining robust when they are unavailable. Comments: Accepted at the USRW Workshop at RecSys 2026 Subjects: Information Retrieval (cs.IR) Cite as: arXiv:2609.25825 [cs.IR] (or arXiv:2609.25825v1 [cs.IR] for this version) https://doi.org/10.48550/arXiv.2609.25825 Focus to learn more arXiv-issued DOI via DataCite (pending registration) Submission history From: Aleksandr Vladimirovich Petrov [view email] [v1] Tue, 22 Sep 2026 07:54:14 UTC (58 KB)
[IR-12] From Offline Proxies to Online Decisions: A Layered Engagement Evaluation Framework for Conversational AI
链接: https://arxiv.org/abs/2609.25408
作者: Xuanyi Li,Vaskar Nath,Hossein Amirkhani,Jay Li,Alex Deng
类目: Artificial Intelligence (cs.AI); Information Retrieval (cs.IR); Machine Learning (cs.LG)
备注: 11 pages main text, 9 pages supplementary material; 2 figures, 25 tables
Abstract:Online A/B experiments are the decision standard for user engagement, but traffic and readout time limit how many conversational-AI changes can be tested. We ask whether an offline signal designed to be computable without treatment-arm user exposure agrees with the outcomes of those experiments. We contribute a reusable construction and diagnosis checklist that treats an offline proxy as a chain of three alignments: behavioral label to product outcome, learned classifier to candidate-assistant behavior, and aggregated offline signal to experiment effect. A companion evaluation protocol audits the whole composite by interval-aware decision agreement, which compares offline and online confidence intervals instead of point estimates, and by within-experiment ranking. The instantiation we evaluate comprises a fixed evaluation suite on which candidate behavior is scored, an engagement classifier trained to predict session/prompt level engagements, and a calibration layer mapping sample-level score differences to online model-level engagement deltas. We then report the audit: 489 paired offline-online contrasts (one candidate arm against its control) from 27 experiments on a deployed multi-turn assistant, spanning model checkpoints to system-prompt tuning. Our primary test uses the 113 contrasts from eight experiments that ran after the map was frozen: on these the composite reaches 81.1% F1, against 34.3% for the raw classifier score it is built on, and makes no wrong-direction calls where that raw score makes 31. Every offline prediction was computed before its experiment ran to prevent overfitting. The evidence supports using the composite to prioritize candidates before scarce experiment traffic is allocated—in our deployment of the experiment, selecting among training checkpoints and tuning system prompts.
[IR-13] ReFilter: Bridging Embeddings and LLM Filtering for Similar Mobile App Retrieval
链接: https://arxiv.org/abs/2609.25306
作者: Buthayna AlMulla,Maram Assi,Safwat Hassan
类目: Information Retrieval (cs.IR)
备注: Accepted at the 89th Annual Meeting of the Association for Information Science and Technology (ASIST 2026)
Abstract:Retrieving similar mobile applications (apps) is essential for researchers, developers, and end-users. Researchers use similarity detection to study app ecosystems and trends, developers for competitor analysis, and end-users for focused app recommendations. Existing approaches rely on embedding-based retrieval, which captures semantic similarity but fails to identify functionally similar apps. To our knowledge, no prior work has applied large language model (LLM)-based filtering to this task, due to the high computational cost of evaluating large numbers of app pairs. To address this gap, we propose ReFilter, a hybrid framework that first Retrieves semantically related candidate apps using embeddings and then applies LLM-based contextual Filtering to identify true functionally similar apps with higher precision. This design balances efficiency and accuracy, achieving an F1-score of 90% for retrieving similar apps. By improving the relevance of app alternatives, ReFilter enables more accurate app comparisons and supports improved ecosystem understanding, competitor analysis, and recommendations.
[IR-14] GroundedGEO: Auditing the Evidence Gap in Generative Search Rankings
链接: https://arxiv.org/abs/2609.25189
作者: Yihan Xia,Huiling Fan,Kangrong Zhong,Taotao Wang
类目: Information Retrieval (cs.IR); Artificial Intelligence (cs.AI)
备注: 14 pages, 6 figures, 12 tables
Abstract:Generative search systems rank products and services for consequential decisions, and publishers can cheaply make candidate text look relevant. Yet evidence status is not a text property but a claim-evidence relation: text-only rankers and defenses cannot separate honest detailed content from fabricated detail, creating an identifiability gap. We audit this gap with an evidence-paired benchmark (50 e-commerce queries, 1,950 cases) and a claim-level reranker, GroundedGEO, that penalizes query-relevant claims lacking support in a supplied packet. Matched rich variants control format and volume; packet twins add attestations at fixed text, while thinned packets withdraw them. On the frozen listwise ranker Qwen2.5-7B, unsupported-rich variants show significant normalized rank gain over clean candidates (+0.065 to +0.092 across claim profiles, Holm-corrected), while supported and neutral controls do not; the effect is model-dependent (marginal on MiMo-v2.5, absent on GLM-5.3-Flash). On a frozen pointwise scorer, oracle evidence labels cut the unsupported-rich top-3 rate from 0.65 to 0.43 (laundering from 0.61 to 0.39) at lambda=40 with zero false suppression; packet twins restore the original rates without changing text. Against a 370-claim human gold, all tested automatic judges fail the preregistered reliability gate, although the best local judge retains 79-100% of oracle suppression with zero measured false suppression on protected arms. Separately, stripping attestation coverage increases false suppression by 0.307. These diagnostic effects identify two limits on the evidence channel: label quality and packet coverage. They do not validate an automatic defense, and interpretation of the adverse human-gold arm remains pending adjudication.
[IR-15] Corporate-Family Resolution Is Not a String-Matching Problem: A Public Benchmark Stratified by Name Visibility
链接: https://arxiv.org/abs/2609.04269
作者: Harshit Gupta
类目: Databases (cs.DB); Information Retrieval (cs.IR); Machine Learning (cs.LG); Logic in Computer Science (cs.LO)
备注: 21 pages, 8 tables. Benchmark, full adjudication log, and code reproducing every number: this https URL
Abstract:Deciding whether two supplier records belong to the same corporate family is a prerequisite for spend consolidation, credit exposure aggregation and sanctions screening. It is usually treated as entity matching, but the tasks differ: a family link connects records that are deliberately different entities, and the evidence often appears in neither record. We introduce CorpFam, a public benchmark of 54,864 candidate pairs over 10,307 corporate families, derived from 6,638,350 US federal award records in which every supplier self-reports its ultimate parent to a government registry. Pairs are stratified by name visibility: whether the names are identical after normalisation, share a distinctive token, or share none. Because strata have positive rates from 10.2% to 97.3%, we report per-stratum recall, base-rate invariant, rather than F1, which is not. The strongest of 5 matchers recovers 100.0% of identical pairs and 4.2% of invisible ones; no method exceeds 4.7% on the latter. The failure begins before matching. Blocking decides which pairs a matcher sees, and we evaluate 7 schemes spanning phonetic keys, attribute keys that ignore the name, and semantic nearest neighbours. None reaches three percent on invisible pairs, and their union recovers 6.8%. 93.2% of these links never enter the candidate set, so no matching-stage improvement can reach them. The links are real: against SEC Exhibit 21 subsidiary schedules, which share no provenance with procurement registration, 64.2% of invisible links are corroborated, against 0.16% under permuted parents and 0.41% against the same parent’s wrong exhibit: two unrelated nulls agreeing to within 0.25 points. Corporate-family resolution is a retrieval problem misfiled as a matching problem; the intervention point is candidate generation, not ranking. The benchmark, adjudication log, and code reproducing every number are released.
人机交互
[HC-0] Does AI Save Time on Product Design? A Randomized Controlled Experiment of AI Prompt-to-Design Workflows
链接: https://arxiv.org/abs/2609.26725
作者: Remy Stewart,Olabode Anise,Andrew Hogan,Augustus Griffin
类目: Human-Computer Interaction (cs.HC); Artificial Intelligence (cs.AI)
备注:
Abstract:AI tools for digital product design now offer prompt-to-design capabilities, allowing designers and their non-designer colleagues to create prototypes through conversational workflows with large language models (LLMs). While these tools promise time savings, experimental evidence in product design remains limited compared with evidence from software engineering. We conducted a randomized controlled trial with 50 product designers and 50 product managers to evaluate prospective time savings from leveraging Figma Make in design work. Participants attempted three standardized design tasks with or without access to Figma Make. Among participants who completed the study tasks, access to Figma Make was associated with approximately 20% shorter completion times, with larger gains among product managers. Our findings suggest that prompt-to-design tools may enable product managers to further contribute to design work, while the benefits for professional designers may be task dependent.
[HC-1] Stepping into the Margins: How Readers Want AI to Generate Footnotes
链接: https://arxiv.org/abs/2609.26673
作者: Piper Vasicek,Courtni Byun,Kevin Seppi
类目: Human-Computer Interaction (cs.HC)
备注:
Abstract:Footnotes can be powerful tools to aid understanding, providing information that augments the reading experience. However, static footnotes cannot address every reader question. Current reading tools allow readers to view curated footnotes, allow personal and social annotation, and link dictionaries to reading material. Many other existing tools and natural language processing (NLP) techniques–such as generative AI, summarization and translation–could be used to address any reader question. However, no one has yet explored which of these features readers actually want. To bridge this gap, we conducted thirteen semi-structured interviews with readers from various backgrounds, followed by a thematic analysis of their responses. We develop themes describing the types of footnotes readers prefer and how to determine the quality of footnotes–specifically focusing on what sources of information a system considers, what the footnotes contain, and how the footnotes are presented to the reader.
[HC-2] Receptiveness Not Sycophancy: Distinguishing Engagement from Deference in Language Models
链接: https://arxiv.org/abs/2609.26579
作者: Calvin Isley,Johann Gaebler,Max Lamparth,Julia Minson,Sharad Goel
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI); Human-Computer Interaction (cs.HC)
备注:
Abstract:A central concern with language models is sycophancy: their tendency to defer to users’ views at the expense of independent substantive judgment. In parallel, work on social sycophancy has focused on behaviors such as validation and positivity that may signal inappropriate deference. Yet the markers of social sycophancy are also characteristic of conversational receptiveness, a construct from social psychology shown to improve interactions across disagreement. We argue that this overlap creates a construct-validity problem for social sycophancy evaluations. Using a popular moral-advice dataset, we find that responses classified as more socially sycophantic are also more receptive. Further, increasing the receptiveness of human-written responses—while preserving their substantive conclusions—causes them to be classified as more socially sycophantic. This tight coupling raises the possibility that social sycophancy evaluations inadvertently penalize desirable behavior. In a preregistered experiment comparing substantively equivalent responses, participants prefer the more receptive responses, expect users to be more likely to listen to them, and are more willing to seek advice from their authors. The same overall pattern persists even among participants who believe the original question asker is in the wrong. Finally, we introduce a simple approach that substantially increases receptiveness without increasing substantive deference, demonstrating that conversational receptiveness and substantive independence can be achieved together.
[HC-3] E3Sense: Head-Confined Multimodal Sensing of Learner Engagement
链接: https://arxiv.org/abs/2609.26569
作者: Sidharth Anupkrishnan,Itir Sayar,Jeongah Lee,Sri Harsha Musunuri,Guan-Ming Su,Madeline Endres,Ravi Karkar,Phuc Nguyen
类目: Human-Computer Interaction (cs.HC)
备注:
Abstract:Engagement-aware learning systems could provide hints or adjust pacing when learners struggle. Prior engagement sensing work distributes sensors across or outside the body rather than consolidating them at one site, or reduces engagement to shared affect or a single dimension (from behavioral, emotional, and cognitive engagement). We introduce E3Sense, a head-worn platform that co-locates electroencephalography, eye tracking, and electrodermal activity to personalize engagement measurement. During a lab study we collected 450 ratings of engagement levels on a five-level ordinal scale while participants watched educational videos. For fifteen held-out participants, E3Sense achieved a within-one-level prediction score of 75.0%, compared with 63.0% for always predicting the most common rating. In an exploratory analysis of 18 participants from the same study who defined engagement, conditioning on learners’ definitions raised the same measure by 6.9 points, from 64.6% to 71.5%. Our work provides a proof-of-concept of a head-site, personalized multimodal sensing of engagement for adaptive educational interfaces.
[HC-4] Bespoke: Generating MOOC-Quality Industry-Personalized Lecture Videos at Scale
链接: https://arxiv.org/abs/2609.26540
作者: Romain Puech,Dewang Kumar Agarwal,Antonio Santamaría Escobar,Dimitris Bertsimas
类目: Human-Computer Interaction (cs.HC)
备注:
Abstract:Lifelong learners from different industries often watch the same recorded lecture, even when they will apply the material in different workplaces. We present Bespoke, a system that takes the transcript of an existing lecture and generates a new video customized to a target industry and duration, with new slides, narration, and charts. From 31 graduate lectures in analytics, machine learning, and optimization, we generated 209 videos for these three industries, plus a generic-audience version of each lecture. Twenty-five experts in the corresponding domains rated 92 videos on a five-point rubric. They judged 87% at or above the rubric midpoint corresponding to ``a standard MOOC lecture’s quality’’ (mean overall score 3.42 out of 5; 48% scored 4 or above). Overall quality was similar across industries, durations, and a 21-lecture held-out set unused while developing the system, at about \ 0.22 API cost per minute of video.
[HC-5] Do Vision Model See Like the Brain? A Comparison Across EEG Encoding Model
链接: https://arxiv.org/abs/2609.26512
作者: Shashank Baghel,Kshitij Dwivedi,Dinesh Singh,Sanjeev Nara
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI); Human-Computer Interaction (cs.HC)
备注:
Abstract:Convolutional neural networks (CNNs) and vision transformers are both used to model the human visual system, but whether the two architectures diverge at a specific point in network depth is unclear. We compared six CNNs and two vision transformers by computing the Pearson correlation ® between each model’s predicted and measured EEG response at every layer or block, in ten participants viewing 200 natural images. For the transformer models, we also tested four token representations, from the classification (CLS) token alone to CLS combined with all patch tokens. CNNs showed strongest correspondence at the earliest layers, weakening at deeper layers, particularly later in the post-stimulus response. Transformers instead sustained strong correspondence at their deepest blocks, though not at their earliest ones. This advantage depended on token representation: pooled representations gave weaker peak correlations (r approx 0.48-0.51) than representations retaining all patch tokens (r=0.640 for CLIP-ViT-B/32, r=0.656 for DINOv2-ViT-B/14). Controlled comparisons showed architecture, not training objective, drove this effect: MoCo-v1 and ResNet-50 (matched architecture) performed nearly identically (r=0.673, 0.670), whereas CLIP-RN50 and CLIP-ViT-B/32 (matched objective) diverged until patch tokens were preserved. We propose that CNN training’s classification bottleneck compresses brain-relevant information at depth, unlike transformers’ self-attention and non-classification objectives. A spatial topography analysis showed a common occipital-dominant pattern across all models, indicating these differences reflect signal strength and persistence rather than distinct brain regions. Patch-preserving transformer representations sustain brain-predictive correspondence where CNNs collapse.
[HC-6] he Ethics of Artificial Intelligence in Military Operations
链接: https://arxiv.org/abs/2609.26507
作者: Nicolas Drapier,Florian Mauberger,Aladine Chetouani,Aurelien Chateigner
类目: Computers and Society (cs.CY); Artificial Intelligence (cs.AI); Human-Computer Interaction (cs.HC)
备注:
Abstract:Deep learning systems now mediate military decisions to use force, yet their internal logic resists inspection, their evaluation practices are gameable, and their deployment fractures accountability across dispersed stakeholders. The ethical challenge posed by these systems is fundamentally epistemic: not just whether autonomous weapons should be permitted to kill, but whether the conditions for responsible human judgment can survive when critical functions are delegated to opaque algorithms. We show that this epistemic condition produces a concrete accountability gap: responsibility diffuses across designers, operators, and policymakers while International Humanitarian Law presupposes capacities for judgment that current AI systems lack. To address this gap, we propose a governance framework that proceduralizes ethical constraints through named accountability roles, adversarial auditing with undisclosed benchmarks, tiered deployment thresholds, and a proposed NATO evaluation standard. Counterfactual analysis of eight documented cases (1988-2025) shows that each governance mechanism addresses a documented class of failure, but no single safeguard suffices in isolation: effective governance of military AI requires not only technical constraints but the institutional infrastructure to keep human judgment meaningful. Subjects: Computers and Society (cs.CY); Artificial Intelligence (cs.AI); Human-Computer Interaction (cs.HC) Cite as: arXiv:2609.26507 [cs.CY] (or arXiv:2609.26507v1 [cs.CY] for this version) https://doi.org/10.48550/arXiv.2609.26507 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
[HC-7] Benchmarking Robots for Everyday Environments: From Lab Experiments to Real-World Operations
链接: https://arxiv.org/abs/2609.26490
作者: Raphael Memmesheimer,Martina Overbeck,Dominik Beyer,Björn Kral,Sabine Bellmann,Sven Schneider,Jan Zimmermann,Anna-Maria Meer,Medina Klicic,Simone Roth,Carolin Straßmann,Alexander Arntz,Marlene Wessels,Johannes Kraus,Paul Schweidler,Tristan Schnell,Christoph Zimmermann,Benedikt Pulver,Wilhelm Stork,Martin Gersch,Sven Behnke,Arne Rönnau
类目: Robotics (cs.RO); Human-Computer Interaction (cs.HC)
备注: 46 pages, 14 figures, 5 tables. Accepted for publication in Frontiers in Robotics and AI (Field Robotics)
Abstract:This study introduces an interdisciplinary framework for benchmarking robots deployed in public environments, addressing the gap between traditional laboratory metrics and real-world benchmarking requirements. We evaluate three distinct robots across diverse use cases - outdoor park cleaning, pedestrian underpass cleaning, and interactive library assistance - each representing unique challenges in public daily life. Over a three-year benchmarking process (2023-2025) comprising seven benchmarking events, a consensus workshop and six on-site evaluations (two per use case), we utilized realistic indoor and outdoor test environments to assess not only technical performance but also the broader implications of deploying robots in unstructured, human-centric settings. An expert panel, spanning robotics, human-robot interaction, safety, and economics, systematically developed and refined an evaluation concept to analyze the transition from laboratory prototypes to operational systems. Our findings highlight critical factors for successful deployment, including task fulfillment, interaction quality, safety, and economic feasibility. This work provides actionable insights for researchers and practitioners aiming to bridge the gap between robotic innovation and real-world applicability.
[HC-8] AI-Generated Email Drafts Shift Culturally Distinctive Communication Styles in Professional Email
链接: https://arxiv.org/abs/2609.26403
作者: Shintaro Sakai,Alice Gao,Yuichi Shoda,Katharina Reinecke
类目: Human-Computer Interaction (cs.HC)
备注:
Abstract:AI assistants that support email composition may shift cultural communication norms, such as the directness typical of low-context cultures like the US versus the indirectness and contextual sensitivity central to high-context cultures like Japan. Yet it remains unknown to what extent people adopt and edit AI drafts inconsistent with their cultural communication norms. We address this through a preregistered within-subject experiment in which Japanese and American participants wrote workplace emails in their native language without AI, with a low-context AI, and with a high-context AI. We found that Japanese participants wrote emails with significantly more high-context markers (politeness, apologies) than Americans. But AI drafts shifted participants’ emails toward the draft’s style, with larger shifts when the draft was culturally misaligned: Japanese drifted most under low-context drafts, Americans most under high-context drafts. These findings suggest AI drafts risk overwriting cultural communication norms unless they adapt to users’ communication styles.
[HC-9] Human-Centricity in Industry 5.0: A Survey of Worker Sensing Adaptive Operations and Human-in-the-Loop Systems
链接: https://arxiv.org/abs/2609.26245
作者: Lara Pereira,Luís Miguel D. F. Ferreira,João Ruivo Paulo
类目: ystems and Control (eess.SY); Human-Computer Interaction (cs.HC)
备注: Accepted at the 17th APCA International Conference on Automatic Control and Soft Computing (CONTROLO 2026)
Abstract:Industry 5.0 (I5.0) repositions manufacturing around a human-centric vision in which cyber-physical systems must adapt to the worker rather than the other way around. Despite significant advances in worker state monitoring technologies, including markerless computer vision, wearable physiological sensors, and machine-learning-based fatigue and cognitive load estimation, a critical integration gap persists: these monitoring capabilities rarely inform real-time operational decisions, and workers almost never participate in the adaptive processes that affect their tasks and workloads. This paper presents a structured survey of the state-of-the-art across three interconnected dimensions: (1) worker state monitoring in industrial settings, encompassing physical fatigue, cognitive load, and ergonomic risk assessment; (2) human factors integration in Operations Management (OM) models, examining how scheduling, task allocation, and production planning incorporate worker-related objectives; and (3) Human-in-the-Loop (HITL) architectures, analyzing the extent to which worker feedback closes the adaptive decision loop. The survey reveals that while each dimension has matured independently, the integration across them remains largely absent. Based on the reviewed literature, no publicly documented and validated system connects real-time sensor-driven worker state estimation with adaptive OM decisions and worker feedback in a unified closed-loop architecture. This gap is identified as the central challenge for realizing the I5.0 vision, and a five-direction research agenda is proposed.
[HC-10] Beyond the Lab: Large-Scale Remote Cybersickness Research in Virtual Reality Using the VERA Platform
链接: https://arxiv.org/abs/2609.26203
作者: Matt Gottsacker,Gerd Bruder,Daniel Zielasko,Alexander Giovannelli,Ali Haskins,Corey Clements,Chloe Beato,John Murray,Robert W. Lindeman,Rui Xie,Jonathan Beever,Nicholas Alvaro Coles,Valerie Jones Taylor,Tabitha Peck,Jeremy Bailenson,Gregory F. Welch
类目: Human-Computer Interaction (cs.HC)
备注: This is a preprint of a publication currently under review at IEEE Transactions on Visualization and Computer Graphics (TVCG)
Abstract:Cybersickness remains a major barrier for the adoption of virtual reality (VR), yet most existing knowledge is derived from laboratory-based studies with relatively small and homogeneous participant samples. In this paper, we investigate whether remote VR studies can produce cybersickness findings comparable to traditional in-lab experiments while enabling larger and more diverse participant populations. Using the Virtual Experience Research Accelerator (VERA), we deployed a remote adaptation of the standardized Cybersicker testbed for cybersickness research and collected data from N=263 participants using their own consumer VR headsets. We compared these results against a previously published in-lab dataset and a demographically matched subset of the remote sample. Across cohorts, cybersickness outcomes were consistent in direction and temporal pattern, including symptom onset trajectories, Fast Motion Sickness Scale (FMS) ratings, and Simulator Sickness Questionnaire (SSQ) responses, supporting the validity of remote cybersickness human-subjects research. Leveraging the larger remote dataset, we additionally examined demographic and individual-difference factors associated with cybersickness. These analyses confirm at scale the effects of sex, sickness susceptibility, and sickness expectation reported in smaller laboratory samples, and they add well-powered evidence on the contested relationship between age and cybersickness. These findings demonstrate that the VERA platform provides a viable means of remotely replicating laboratory-based studies and highlight its ability to support large, demographically diverse participant samples. This work establishes a validated methodological foundation for future large-scale remote studies of cybersickness and other VR research topics.
[HC-11] RACE: Transparent Retrieval for Abstract Concept Evaluation
链接: https://arxiv.org/abs/2609.26168
作者: Joseph Bingham
类目: Human-Computer Interaction (cs.HC); Computer Vision and Pattern Recognition (cs.CV)
备注: 12 pages, 3 figures
Abstract:Recent work reports that vision–language models (VLMs) struggle to establish and maintain stable reference in repeated reference games. Rather than ask which VLM does best, we ask a more basic question: do you need a large pretrained VLM for this at all? On grounding a single director utterance to one of twelve tangram silhouettes, we compare six off-the-shelf VLMs against a transparent baseline that uses \emphno learned visual representation: classical SIFT keypoint matching and a signal-quality index over retrieved images. On identical trials, the transparent baseline matches the strongest VLM (SigLIP-large) and significantly outperforms the other five, including every CLIP and OpenCLIP variant. The baseline additionally retrieves external images, so this is not a matched-information comparison; what it shows is that a learned \emphvisual representation is not the bottleneck for this task: given retrieved images, a shape-appropriate classical similarity suffices. Along the way we find that abstract-grounding ability varies widely across VLMs (15–39% top-1; chance 8.33%, humans \approx 77–80%), so the weakness is model-specific rather than intrinsic to contrastive pretraining; on the 1,013-shape KiloGram benchmark the pattern generalizes for CLIP, with per-shape difficulty tracking human shape-nameability. The pipeline is a classical, inspectable alternative rather than a learned one. We close by sketching how an explicit, inspectable representation of listener-side pact state could carry this approach into interactive multi-turn reference, which we leave to future work. Code available in supplementary material.
[HC-12] Designing Task-Induced Arousal: A Multimodal Stress Induction Method for Interactive Experiments
链接: https://arxiv.org/abs/2609.26156
作者: Morten Roed Frederiksen
类目: Human-Computer Interaction (cs.HC); Robotics (cs.RO)
备注: Preprint submitted version, 8 pages, 4 figures
Abstract:HCI and HRI studies often require short, repeatable arousal manipulations that can run while participants continue interacting with a device or robot. These experiments are often challenged by the need to induce arousal in settings that still resemble real interaction. Participants must continue using a device, touching a robot, or producing sensor data while the manipulation unfolds. We aimed to develop a compact and repeatable way to induce controlled task-related arousal during interactive experiments by combining a lightweight browser-based pacing task, escalating timing demands and urgency cues, and a concurrent physical hotwire-style challenge. In an A-B-A within-participant study, the induction condition significantly increased mental demand, temporal demand, effort, frustration, and SAM arousal (all p .001), while perceived performance decreased (p .001). GSR peak rate increased relative to both calm conditions (p = .030) and escalated over time (p .001). Grip variability also increased (p = .036), as did release speed (both p .001), while valence remained above the scale midpoint. These results provide initial evidence that the combined procedure induces controlled, relatively high-valence task-related arousal and may serve as a reusable experimental tool for future human-computer and human-robot interaction studies.
[HC-13] Instructional Governance by Design: A Framework for AI in Computing Education
链接: https://arxiv.org/abs/2609.26098
作者: Ethan Dickey
类目: Computers and Society (cs.CY); Human-Computer Interaction (cs.HC)
备注: 7 pages, 3 tables
Abstract:As generative AI permeates computing instruction, the emergent design challenge is to configure each tool’s pedagogical role, authority, and accountability for the instructional work it performs. We argue for instructional governance by design: governance should be encoded in a teaching tool’s interaction model, constraints, and workflow. We introduce a multidimensional framework that characterizes AI teaching tools through (1) pedagogical grounding, (2) AI instructional authority, (3) human accountability and control, (4) learner agency and cognitive engagement, (5) context specificity and boundary setting, and (6) evaluation visibility and revision. These dimensions yield governance profiles that help educators align tools with specific purposes and educational stakes. We develop the position through a comparative analysis of a portfolio of AI teaching tools across computing and first-year engineering: rubric-anchored GTA simulations, reflection-oriented code companions, staff-reviewed forum-response systems, TA-supervised diagram generators, and course-specific code-style coaches. These cases show how common instructional functions call for different combinations of rubrics, learning-theory commitments, approval gates, supervision, and course-specific constraints. We further apply the framework to selected published tools to demonstrate its use beyond a single institutional portfolio. From these cases, we identify reusable design questions for aligning governance with instructional stakes, human capacity, and intended learning processes. This position reframes responsible AI integration as a curricular and interaction-design challenge and offers a common vocabulary for tool builders, instructors, and researchers to design, compare, and evaluate AI-mediated learning environments.
[HC-14] Faithful Faithfulness Evaluations: Challenges Pitfalls Learned from a Breast MRI Case Study MICCAI
链接: https://arxiv.org/abs/2609.25978
作者: Peachapong Poolpol,Henrik H. J. Detjen,Eike Petersen
类目: Computer Vision and Pattern Recognition (cs.CV); Human-Computer Interaction (cs.HC); Machine Learning (cs.LG); Image and Video Processing (eess.IV)
备注: Accepted at MICCAI iMIMIC Workshop 2026
Abstract:Saliency maps are widely used to explain deep learning predictions in medical imaging, yet visually plausible explanations do not necessarily reflect a model’s true decision process and may therefore mislead clinicians. We investigate this problem using a Vision Transformer-based breast MRI classifier trained on the ODELIA Breast MRI Challenge dataset and evaluate multiple saliency methods, including Last-layer Attention, Attention Rollout, Grad-SAM, Gradient Attention Rollout, GMAR, Grad-CAM, and HiResCAM. Our study highlights two often-overlooked challenges in perturbation-based faithfulness evaluation. First, method rankings depend strongly on the perturbation strategy, varying across intensity-based perturbations and transformer-based attention masking. Second, benchmarking saliency methods requires distinguishing between class-specific and class-agnostic explanations. To enable fair comparisons, we introduce non-class-specific variants of gradient-based methods and evaluate both settings separately. Across protocols, Grad-CAM and Gradient Attention Rollout consistently emerged as the strongest class-specific methods, although their relative ranking depended on the evaluation design. These findings expose important limitations of current saliency-based explainability approaches and highlight the need for more robust and standardized evaluation frameworks for trustworthy clinical AI systems.
[HC-15] How Its Made: Uncovering Detection Engineering Processes for Network Intrusion Detection Rules
链接: https://arxiv.org/abs/2609.25901
作者: Koen T. W. Teuwen,Emmanuele Zambon,Luca Allodi
类目: Cryptography and Security (cs.CR); Human-Computer Interaction (cs.HC)
备注:
Abstract:Many Security Operations Centers rely on signature-based Network Intrusion Detection Systems like Suricata, yet detection rule engineering remains understudied. We investigate this process by introducing SuriCap, a platform for rule engineering exercises, and hosting CTF-style workshops where 60 participants, trained MSc students, and experienced SOC professionals, created rules for four scenarios. Participants produced 3146 valid rules, enabling analysis of their methods, performance, and iteration patterns. Surprisingly, prior experience had limited impact on rule quality, suggesting that less experienced engineers can produce rules comparable to experts. We also observed challenges in generalizing rules beyond available tests, underscoring the need for sufficient labeled data. From our study, we identify three phases and a common pattern in rule engineering, offering SOC managers insights to improve their processes and expectations of engineer expertise.
[HC-16] he Impact of Deep Care Isa on Reducing Musculoskeletal Disorders and Enhancing Productivity Among Office Employees: A Comprehensive Study
链接: https://arxiv.org/abs/2609.25883
作者: Quanmin Liang,Junjie Yang,Mohammad Ali Nasseri,Daniel Zapp
类目: Human-Computer Interaction (cs.HC)
备注:
Abstract:Prolonged sedentary behavior, a pervasive issue in modern workplaces, has been closely linked to musculoskeletal disorders (MSDs) and reduced productivity. This study evaluates the effectiveness of Deep Care Isa, an advanced digital health assistant, in addressing these challenges. Utilizing data from over 2,300 participants across 50 corporations, the study demonstrates significant improvements in ergonomic practices, physical activity, hydration habits, and overall productivity, with notable reductions in MSD-related sick leave. The findings highlight the role of innovative ergonomic interventions in enhancing employee well-being and organizational efficiency. Comprehensive statistical analysis underscores the reliability and practical significance of these outcomes.
[HC-17] Harnessing LLM s Without Surrendering Control: Delegation Boundaries in Visual Data Storytelling Authoring IEEE-VIS2026
链接: https://arxiv.org/abs/2609.25700
作者: Zhuojun Jiang,Yuki Ueno,Chris Bryan
类目: Human-Computer Interaction (cs.HC)
备注: Accepted as a Short Paper at IEEE VIS 2026
Abstract:Despite the emergence of large language models (LLMs) for visual data storytelling workflows, there are open questions about how authors decide what activities or tasks to entrust to them and what should be “protected” or maintained under human control. To investigate this, we interviewed a cohort of 12 expert visual data storytellers. Our analysis shows that participants rarely treated LLMs as autonomous storytellers. Instead, they tend to selectively delegate execution-oriented tasks to LLMs while retaining control over activities that shape narrative intent and story meaning. Our findings show that LLM assistance is most productive after human seeding and constraint-setting, and that it shifts labor from production to verification. We discuss design implications for boundary-aware authoring tools, data-grounded generation, low-fidelity ideation, and reporting practices for LLM-based visualization research. Supplemental materials for this paper are available at this https URL.
[HC-18] SurgGraph: Quantitative Laparoscopic Video Understanding via Geometry-Grounded Scene Graphs
链接: https://arxiv.org/abs/2609.25651
作者: Jingying Wang,Rosiana Natalie,Marquise D Singleterry,Filippos Bellos,Brian George,Gurjit Sandhu,Jason J Corso,Anhong Guo,Vitaliy Popov,Xu Wang
类目: Human-Computer Interaction (cs.HC)
备注:
Abstract:Surgical videos are a primary resource for teaching trainees anatomy, tool usage, and procedural skills. Yet learning from them at scale requires systems that understand surgical scenes. Existing approaches fall short: vision-language models lack fine-grained domain reasoning, task-specific models do not generalize, and prior scene graphs omit clinically meaningful detail. We present SurgGraph, a training-free pipeline that generates quantitative scene graphs from surgical videos. Operating on segmentation masks and depth maps, SurgGraph encodes each clinically meaningful relation (attachment, occlusion, separation, tool actions) as a subject, verb, object, value tuple whose numeric value quantifies the relation’s extent over time. Technical evaluations show more precise scene understanding than state-of-the-art surgical VLM baselines. We then build SurgGraphQA, a proof-of-concept learning application that retrieves meaningful and boundary-case exemplars and generates visual explanations and feedback. A study with 17 medical students and 2 resident surgeons shows significant learning gains, demonstrating its educational value.
[HC-19] SurgGaze: Implicit Calibration for Accurate Gaze Analysis in Operating Rooms with Wearable Eyetrackers
链接: https://arxiv.org/abs/2609.25612
作者: Jingying Wang,Rosiana Natalie,Keyuan Hu,Wenqian Xu,Brian George,Vitaliy Popov,Anhong Guo,Xu Wang
类目: Human-Computer Interaction (cs.HC)
备注:
Abstract:Accurate gaze tracking is essential for understanding surgeons’ visual attention and cognitive processes during laparoscopic surgery, yet wearable eye trackers produce large errors systematically correlated with ground-truth gaze locations, as demonstrated in Study 1. We introduce SurgGaze, an implicit calibration method that corrects these errors using high-confidence surgical moments. Building on evidence that surgeons’ gaze converges near the tool-tissue contact point (TTCP) during dissection, SurgGaze uses TTCP as a surrogate for true gaze to construct training pairs. We evaluate SurgGaze in a simulated operating room trial and an authentic operating room case study. In simulation, SurgGaze reduced gaze estimation error by 40.6%, significantly outperforming conventional 9-point explicit calibration. The case study showed that these moments provide reliable training data and that calibrated gaze improves interpretation of surgeons’ attention beyond numeric error reduction. These findings demonstrate that structured behavioral signals can enable implicit calibration for gaze tracking in complex real-world settings.
[HC-20] Deflecting the Value Compass: Interacting with Large Language Models Temporarily Shifts Human Value Priorities Toward Personal Focus
链接: https://arxiv.org/abs/2609.25586
作者: Hasibur Rahman,Malak Sadek,Smit Desai
类目: Human-Computer Interaction (cs.HC); Artificial Intelligence (cs.AI)
备注:
Abstract:Large language models increasingly support decisions where values are in tension, yet little is known about whether interacting with them changes which values users prioritize. In a preregistered study, 200 U.S. adults interacted with ChatGPT, Claude, or Gemini as a thinking partner or read fixed AI-generated considerations. The prompt asked LLMs to support reasoning without recommending a decision and named no values. Participants advised people facing real dilemmas and completed parallel PVQ-RR forms before, immediately after, and one task later. Each LLM condition temporarily shifted value priorities toward personal focus relative to the control (d=0.37-0.51), primarily through increased Self-Enhancement. Participants’ advice retained words and meaning from their exchanges. Thus, a brief LLM interaction that neither targets values nor seeks to persuade can reorient values active during judgment without detectable convergence in value directions or advice.
[HC-21] When the Strike Zone Becomes Algorithmic: Umpire Judgment and Player Challenge Decisions under AI Review
链接: https://arxiv.org/abs/2609.25525
作者: Kichang Lee,Gyeongmin Han,Sungmin Lee,JeongGil Ko
类目: Human-Computer Interaction (cs.HC); Computers and Society (cs.CY)
备注: 31 pages, 10 figures, 1 table
Abstract:The Automated Ball-Strike challenge system that Major League Baseball adopted in 2026 offers a distinctive setting for studying human AI interaction in which umpires make every ball and strike call, while players can selectively ask an automated system to publicly overturn those decisions. We analyze 4,114,256 called pitches from 2015 through 2026 and 8,447 challenges from the 2026 season to examine how algorithmic review reshapes umpire judgment and player behavior. We study where umpires placed the effective strike zone boundary, how consistently they applied that boundary, how they responded to overturned calls, and which calls players chose to challenge. In 2026, the effective called boundary shifted toward the automated strike zone beyond the trajectory observed in prior seasons, while the consistency of that boundary largely continued its existing trend. Following an overturned call, umpires temporarily adjusted subsequent decisions near the corrected boundary, although these effects did not consistently persist into the next game. Count dependent variation in calling remained, while differences associated with player status narrowed. Players, meanwhile, left many overturnable calls unchallenged and appeared to base challenge decisions more strongly on immediately observable evidence than on the precise geometry of the automated zone. Together, these findings show that selective AI review does more than correct individual errors. It reshapes human judgment, adaptation, and strategic behavior around an algorithmic authority.
[HC-22] Conduct Under Pressure: What Sixty Language Models Do When a User Pushes
链接: https://arxiv.org/abs/2609.25447
作者: Tapan Parikh
类目: Computation and Language (cs.CL); Human-Computer Interaction (cs.HC)
备注: Code, data and labels: this https URL
Abstract:We study what LLMs do when a user applies pressure in an uncomfortable situation: a user insists, begs, flatters or grieves, and the model gives up a correct fact, writes a document it should refuse, or cheers a plan that will cost the user money. We send frozen multi-turn scenes, identical for every model regardless of the reply, to 60 models from 13 vendors, and label each transcript with a codebook built by open coding and then frozen: a trajectory (the model held its position or folded) and a manner (how it held or folded). Two findings separate. Whether a model holds tracks its generation, meaning how recent it is: fold rate correlates with a public capability index at Spearman -0.64, with little vendor effect. How it holds tracks the vendor: six of the 17 manner codes sort by vendor at permutation p = 0.001, corrected across the codebook. We report four vendor profiles on the codes that cleared reliability. We also ask which parts of the labeling need a person. Six LLM coders from three vendors apply the codebook more consistently than three human coders do (Krippendorff’s alpha 0.66 against 0.46), agree with the codebook’s author on trajectory at kappa 0.84 to 0.91 on transcripts the codebook’s examples never touched, and match an adjudicated human reference at 0.83. Blind machine readings recover the codebook’s categories but cannot tell which of them a second reader would apply the same way. We conclude that for behavior a non-specialist can judge, the human contribution is authoring and bounding the codes and owning a small reference, not producing labels at volume. Comments: Code, data and labels: this https URL Subjects: Computation and Language (cs.CL); Human-Computer Interaction (cs.HC) Cite as: arXiv:2609.25447 [cs.CL] (or arXiv:2609.25447v1 [cs.CL] for this version) https://doi.org/10.48550/arXiv.2609.25447 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
[HC-23] How Early Can You Tell? Early Eye Gaze Dynamics and Cybersickness Progression in Virtual Reality
链接: https://arxiv.org/abs/2609.25422
作者: Nevzat Umut Demirseren,Isayas Berhe Adhanom
类目: Human-Computer Interaction (cs.HC)
备注: 6 pages, 3 figures
Abstract:Cybersickness remains one of the primary barriers to prolonged and comfortable virtual reality (VR) use, yet the temporal development of cybersickness remains poorly understood because its onset and progression vary substantially across individuals. Eye tracking provides a continuous and unobtrusive measure of user behavior, making it well suited for investigating how users respond throughout VR exposure. However, existing work has mainly focused on aggregate gaze measures or post-exposure assessments, leaving the relationship between early eye-movement dynamics and discomfort progression underexplored. In this paper, we investigate whether early eye-movement dynamics characterize the subsequent progression of cybersickness. Using trajectories of fixation duration, fixation dispersion, saccade amplitude, saccade velocity, and gaze eccentricity computed over the first one to five minutes of VR exposure, we evaluate their association with subsequent discomfort progression and compare early gaze dynamics between participants with low and high post-exposure cybersickness severity. Our results show that early fixation dispersion is consistently associated with subsequent discomfort progression, whereas no significant differences are observed between cybersickness severity groups. These findings suggest that early fixation dispersion can characterize the progression of discomfort rather than the final cybersickness severity. Overall, this work provides new insight into the temporal relationship between early eye-movement behavior and cybersickness progression, contributing to the understanding of gaze dynamics during immersive experiences.
[HC-24] Beyond Task Performance: Lessons Learned from Evaluating an Exploratory VR Interaction Technique
链接: https://arxiv.org/abs/2609.25414
作者: Nevzat Umut Demirseren,Corey Pittman,Isayas Berhe Adhanom,Karthikeyan Umapathy,Kevin Pfeil
类目: Human-Computer Interaction (cs.HC)
备注: 6 pages, 10 figures
Abstract:Dense virtual environments present significant challenges for object selection and manipulation, motivating the development of novel interaction techniques. This paper presents RodCast as an exploratory case study to investigate explicit trajectory visualization for interaction in dense virtual environments. We conducted a within-subjects user study comparing RodCast with Go-Go Hand and FlowerCone across three representative interaction tasks using objective performance measures, subjective evaluations, and qualitative feedback. The results revealed that the proposed implementation incurred performance costs on more demanding manipulation tasks, while qualitative feedback suggested that participants attributed these challenges to the complexity of the bimanual control scheme. Additionally, subjective evaluations and qualitative feedback identified benefits in spatial awareness and target accessibility that were not fully reflected by conventional performance measures. Together, these findings highlight the importance of control simplicity in interaction technique design and emphasize that incorporating qualitative feedback into the evaluation process is essential for distinguishing implementation limitations from the potential of the interaction concept. Collectively, these lessons provide guidance for the design and evaluation of future interaction techniques.
[HC-25] Learning from Humans for Proactive Assistance in Human-Robot Collaborative Transport
链接: https://arxiv.org/abs/2609.25351
作者: Elvin Yang,Christoforos Mavrogiannis
类目: Robotics (cs.RO); Human-Computer Interaction (cs.HC); Machine Learning (cs.LG)
备注:
Abstract:We focus on human-robot collaborative transport, a challenging task of broad relevance spanning logistics, manufacturing, and the home, in which a user and a robot work together to relocate a large or heavy object. To act as an effective partner, the robot should reduce the user’s effort by contributing to efficient relocation of the object while remaining physically responsive to them. Prior work often addresses these capabilities separately, producing robots that may move the object efficiently but resist user input, or accommodate the user but depend on continuous guidance. Our key insight is that obstacle-constrained collaborative transport requires integrating predictions of human collaborative behavior with compliant robot control. To this end, we introduce PROACT, a framework for human-robot collaborative transport that incorporates anticipation into compliant whole-body control through a learned model of human collaborative behavior. Trained on a large-scale, real-world dataset of dyadic human transport demonstrations, our transformer architecture distills collaborative behavior into predictions of future object motion. Across 108 real-world trials with a 9-DoF mobile manipulator, PROACT reduces mean interaction work by 59.2% and 20.4%, and mean completion time by 12.9% and 6.9%, relative to compliance-only and MPC baselines, respectively. Footage from our experiments can be found at this https URL.
[HC-26] “I Talked an AI Chatbot So Whats Next?” How U.S. Young Adults Imagine Responsible AI for Emotion Coping
链接: https://arxiv.org/abs/2609.25311
作者: Jiaying Liu,Nimra Ishfaq
类目: Human-Computer Interaction (cs.HC)
备注:
Abstract:Emotion coping is inherently relational, unfolding through interactions with friends, family, professionals, and communities. Yet AI chatbots are largely designed around a user–AI dyad. Learning from the ethics of care, we examine how AI chatbots shape the relational conditions of emotion coping. We conducted a scenario-based study with 17 U.S. young adults across four emotion coping scenarios. Participants identified eight roles through which AI could support relational conditions, alongside three challenges: flattening distinct relational conditions, discouraging reciprocity, and shifting relational labor onto users. This study contributed a relational perspective of responsible AI in emotion coping. We argue that responsible AI should respond to individuals’ situated relational conditions rather than provide general-purpose support. We further identify two design principles: fostering reciprocity by supplying materials to engage with others, and strengthening emotional self-efficacy. Together, we position responsible AI as AI in the loop of human relationships.
[HC-27] Can LLM s identify and repair ruptures? Comparison between clinician practices and LLM behaviors
链接: https://arxiv.org/abs/2609.25287
作者: Jeongah Lee,Joy Qiuyue Zhong,Drishti Goel,Violeta J. Rodriguez,Dong Whi Yoo,Koustuv Saha,Ravi Karkar
类目: Human-Computer Interaction (cs.HC)
备注:
Abstract:Ruptures represent common albeit critical moments in interaction where relational alignment breaks down, making them essential for evaluating AI where trust and engagement matter most. In a scenario-driven empirical study, we examined the performance of three LLMs at identifying and resolving ruptures across 21 mental health conversations and 22 experts’ evaluation of the strategies. For identification, LLMs relied on explicit linguistic cues within single turns whereas experts integrated implicit, relational, and contextual information across the conversation. For resolution, LLMs tended to produce more directive and scripted responses whereas experts adopted process-oriented strategies such as validation, open-ended exploration, and psychoeducation. Overall, LLMs showed higher agreement with predefined labels in identification, but not in resolution where experts rated their responses only moderately effective, with consistent limitations in timing, depth, and contextual sensitivity. We discuss implications for the design of mental health conversational agents emphasizing relational awareness, pacing, and human-in-the-loop support.
[HC-28] A Case Study in Accessible Redesign of a Wastewater Dashboard IEEE-VIS2026
链接: https://arxiv.org/abs/2609.25273
作者: Tingying He,Jake Wagoner,Md Rahat-uz- Zaman,Md Dilshadur Rahman,Willy Ray,George G. Vega Yon,Matthew Samore,Alexander Lex,Paul Rosen
类目: Human-Computer Interaction (cs.HC)
备注: In Proceedings of the AccessViz Workshop at IEEE VIS 2026
Abstract:Public health dashboards communicate data that can inform important decisions, but they often raise accessibility challenges. We present a case study redesigning the Utah Wastewater Surveillance System dashboard to improve accessibility and usability across desktop and mobile settings. The redesign was informed by WCAG and developed iteratively with Utah DHHS collaborators, and we collected feedback on the final design from an external blind researcher. Our case study highlights that accessible dashboard design requires aligning accessibility guidelines with user needs, stakeholder workflows, and technical constraints. It also suggests that simple, targeted technical solutions tailored to the existing environment can provide practical value for dashboard redesign in government contexts.
[HC-29] How Children Design and Reason about Trustworthy AI Chatbots
链接: https://arxiv.org/abs/2609.25244
作者: Deniz Ozturk,Jiayu Li,Daksh Pratap Singh,Yasitha Rajapaksha,Fasika Melese,Bahare Riahi,Shiyan Jiang,Qiao(Georgie)Jin,Joey Huang,Veronica Cateté,Tiffany Barnes,Xiaoyi Tian
类目: Human-Computer Interaction (cs.HC); Artificial Intelligence (cs.AI)
备注:
Abstract:Children increasingly interact with AI chatbots, making trust calibration essential to AI literacy. Prior research has examined children’s trust in AI mainly as users evaluating systems built by others, rather than as designers of their own chatbots. We developed a chatbot-building environment with adjustable trust-relevant traits (e.g., confidence, transparency, formality, assertiveness), rules, and persona. We conducted mixed-methods study with 115 learners (ages 8-18) who made 119 chatbots. We examined how children configured their chatbots, reasoned about trustworthiness, and how closely chatbot behavior aligned with their designs. Younger students (age 10-13) set significantly higher confidence than older students (age 14-18), and some deliberately built chatbots that gave wrong answers on purpose, yet still called them trustworthy, arguing that a chatbot does what it was built to do. Younger students equated trust with purpose-fulfillment, while older students linked it to transparent, calibrated design. Students also calibrated academic chatbots to be more transparent and formal than hobby chatbots. We identify seven design dimensions describing what children believe makes a chatbot trustworthy, and discuss implications for AI literacy tools.
[HC-30] MolExplain: An Interactive Tool for Explainable Molecular Property Prediction
链接: https://arxiv.org/abs/2609.25355
作者: Pirm Dhararag,Dylan Cashman
类目: Chemical Physics (physics.chem-ph); Human-Computer Interaction (cs.HC)
备注:
Abstract:The application of machine learning to molecular property prediction has become increasingly prevalent in drug discovery, yet most models operate as black boxes, returning a prediction without revealing which structural features drive it. MolExplain addresses this gap by combining property prediction with sub-structure level visual explainability in an interactive web interface. The system featurizes molecules as Morgan fingerprints, classifies them using a trained XGBoost model, and applies SHAP attribution to produce a smooth heatmap overlay indicating which regions of the molecule contribute for or against the predicted property. Applied to cyclic peptide membrane permeability, the tool’s attribution independently recovers the known role of backbone N-methylation in improving passive membrane diffusion, consistent with established chemistry. While demonstrated on cyclic peptides, the framework is designed to generalize to other molecular properties, positioning MolExplain as a platform for interactive, explainability-driven molecular design.
计算机视觉
[CV-0] ϕ-RIE: From Photorealistic Reconstruction to Interactive Environments
链接: https://arxiv.org/abs/2609.26795
作者: Runyi Yang,Deheng Zhang,Xiaoye Wang,Kanzhi Wu,Lei Sun,Ajad Chhatkuli,Kunyu Peng,Luc Van Gool,Danda Pani Paudel
类目: Robotics (cs.RO); Computer Vision and Pattern Recognition (cs.CV); Graphics (cs.GR)
备注: 8 pages, 6 figures
Abstract:3D Gaussian Splatting (3DGS) can reconstruct a captured scene photorealistically, but the resulting representation does not by itself support physical interaction. Robot simulation instead requires object-level change, \textiti.e., objects must move independently, make contact, and reveal previously occluded surroundings. This gap arises because object appearance may remain entangled with the background, while hidden object geometry and occluded background content may be unobserved. To address this challenge, we present \phi-RIE, a Gaussian-native pipeline that converts selected objects into movable simulator assets while preserving the remaining reconstruction. Our key observation is that asset construction and source removal should be coupled, \textiti.e., one object identity should define the movable asset and the scene content to remove and complete. Accordingly, Scene Observation supplies shared evidence to Coupled Scene Construction, which creates registered assets and completed background Gaussians for simulator-driven rendering in an Interactive Environment. This coupling preserves unedited Gaussians while aligning visual and physical state. On 50 ScanNet++ scenes, evidence-based selection and registration retry increase matched F1 at 20,mm from 0.336 to 0.383 at fixed retention. Further tests demonstrate asset executability, manipulation gains over a single-generator baseline, and the visual cost of conversion. Together, these results demonstrate that \name\ enables interactive scene conversion.
[CV-1] HARMONY: Hierarchical Agent ic Reasoning for MONocular Image-to-Scene Synthesis
链接: https://arxiv.org/abs/2609.26793
作者: Shufan Sun,Chen Wang,Enxin Song,Jiatao Gu,Lingjie Liu
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: Project Page: this http URL
Abstract:Compositional 3D scene reconstruction has recently been explored from two directions: agentic reasoning that provides semantic understanding of spatial relationships but lacks precise alignment with input images; and visual geometry foundation models that predict dense point maps from input images but the reconstruction quality is limited. Therefore, recovering a complete 3D scene from a single monocular image with accurate inter-object relationships and high-fidelity reconstruction quality remains challenging. In this paper, we present HARMONY, a hierarchical chain-of-thought framework that leverages both agentic reasoning and visual geometry foundation. Given an image of an indoor scene, starting from an empty 3D floorplan, HARMONY first calibrates the camera against the reference image to establish a semantically-grounded spatial frame, then uses agentic VLM reasoning to recover the 3D room layout and an initial placement order. It then places the objects in a hierarchical order, from wall-mounted elements, free-standing furniture, to dependent decorations on top of furniture. We also use depth-first traversal for furniture so each placement conditions on previously resolved structure and a reflective feedback loop to avoid error accumulation. After each object placement by VLM, we use the point cloud estimations to perform geometry-based refinement so that the rendered image aligns better with the input. HARMONY can produce 3D scenes that are semantically consistent and perceptually aligned with the reference image, extending single-image compositional reconstruction to complex indoor scene images. Experiments on synthetic and real-world images demonstrate that HARMONY outperforms the evaluated reconstruction baselines, while qualitative comparisons with GPT-6 Astra suggest more faithful object arrangements and better preservation of scene details.
[CV-2] DreamStream: Towards Policy-Oriented Generative Simulation for End-to-End Driving
链接: https://arxiv.org/abs/2609.26792
作者: Ziyang Leng,Sicheng Mo,Seth Z. Zhao,Haoyuan Cai,Yu Zeng,Rowan McAllister,Bolei Zhou
类目: Robotics (cs.RO); Computer Vision and Pattern Recognition (cs.CV)
备注: Accepted to CoRL 2026. Project page: this https URL
Abstract:Faithfully evaluating end-to-end driving policies in simulation requires observations that are not merely photo-realistic, but preserve the scene features a policy relies on to make decisions. Existing platforms, however, exhibit a sim-to-real visual gap that corrupts policy perception, undermining their ability to assess a policy’s closed-loop decision-making. To this end, we propose DreamStream, a generative, closed-loop simulator that achieves policy-oriented fidelity using a simulator-grounded autoregressive video model. Our video model is distilled from a large pretrained video model via traffic layout guidance, varying visual appearance while preserving policy-relevant features such as scenario layout and the temporal consistency of dynamic objects. We further observe that perceptual metrics like FID misrank how well these features are preserved. To tackle this, we introduce FD \pi , a new multi-representation metric that measures the sim-to-real gap as the Fréchet distance over scene-context features from public E2E policies. Under FD \pi , DreamStream improves over the strongest prior closed-loop simulator by 1.6\times on nuScenes and 4.7\times on NAVSIM, and induces the least perturbation to policy’s perceptual observability. Based on DreamStream, we construct Navhard-CL benchmark, which turns non-reactive real-world benchmark NAVSIM into interactive testing environments with adversarial driving behaviors and weather variations. This benchmark exposes many failure modes of driving policies, such as scorer bias and lack of recovery behaviors, that prior closed-loop benchmarks overlook. Code and data are available at this https URL.
[CV-3] StableVQ: Practical Guidelines for Stable Vector-Quantized Tokenizer Training
链接: https://arxiv.org/abs/2609.26774
作者: Bao Tang,Jiahao Guo,Haoxiang Cao,Wenyu Liu,Changqian Yu,Kun Gai,Xinggang Wang
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: Project page: this https URL
Abstract:Vector Quantization (VQ) is fundamental to discrete visual tokenizers that power modern autoregressive and masked image generation models. While recent shared-projection codebook methods have substantially advanced codebook utilization, training stability remains a critical and underexplored challenge. We argue that the root cause lies in the entanglement of the Encoder–Decoder and Codebook training: because neither module can reliably fulfill its own responsibility in isolation, the system can only function when the two subsystems happen to cooperate—a fragile condition that breaks down precisely when training is most stressed. We propose StableVQ, which revisits the proper learning objective of each module and resolves the problems that arise when each is trained to fulfill its own role independently. Concretely, (1) Dynamic STE corrects the instability in the Encoder’s learning objective, enabling it to robustly optimize the reconstruction space under discrete regularization even when codebook utilization is low. (2) Region VQ Loss reconceives the Codebook’s learning objective so that it can independently guarantee full tracking of the encoder output distribution, without relying on encoder oscillations to drive activation. (3) Decoupled Schedule recognizes that the distinct responsibilities of the Encoder–Decoder and the Codebook demand distinct optimization dynamics, and assigns each an independent learning rate schedule to ensure robust system-level behavior. Built on top of shared-projection codebooks, StableVQ is lightweight and introduces no learnable parameters. Experiments on ImageNet demonstrate consistent improvements in training stability, codebook utilization, and reconstruction quality across diverse codebook sizes and initialization settings.
[CV-4] FleXray: Universal Clinical X-ray Segmentation
链接: https://arxiv.org/abs/2609.26756
作者: Victor Ion Butoi,Vivek Gopalakrishnan,John V. Guttag,Adrian V. Dalca,Neel Dey
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
备注: 35 pages, 12 figures, 10 tables. Code, models, data, and a browser-based demo at this https URL
Abstract:X-ray is medicine’s most widely used imaging modality, yet remains among its least quantitative. Unlike volumetric modalities like CT or MRI, X-ray collapses 3D anatomy into a 2D projection, causing structures to overlap and anatomical boundaries to be ambiguous, even to experts. As a result, labeling X-ray databases for training general-purpose segmentation systems is impractical, leaving morphometric and functional X-ray analysis confined to narrow anatomical regions and applications. To this end, we present FleXray, a generalist model for anatomical segmentation across the entire body in clinical X-rays. Instead of curating large, manually annotated X-ray datasets, we build a scalable, physics-based generative X-ray data engine. Using existing 3D whole-body CT segmentation datasets and generative image-editing models, we simulate fully-annotated 2D X-rays with diverse appearances, physiological properties, and imaging geometries. Trained on these simulations, FleXray accurately segments 60 anatomical structures across unseen research datasets and in-the-wild X-rays. We further show that FleXray makes X-rays directly amenable to quantitative analysis, enabling automated measurements for disease grading, robust navigation during X-ray-guided interventions, and data-efficient learning of pathological targets. We release the model, code, a full-body X-ray segmentation dataset, and a local, easy-to-use browser-based tool at this https URL .
[CV-5] Evaluating the Semantic-to-Geometric Gap in Adversarial Defenses Against Vision-Language Model-Based Plagiarism
链接: https://arxiv.org/abs/2609.26733
作者: Christopher Burger,Christina Trotter,Joseph Carlisle,Charles Walter
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: 10 Pages, 3 figures, 2 tables
Abstract:The rapidly advancing capabilities of vision-language models (VLMs) present a systemic challenge to academic integrity. VLMs now allow students to bypass meaningful engagement by capturing and submitting graphical problems as singular images, a practice we define as trivial plagiarism. To provide educators with actionable data on VLM limitations, we investigate the efficacy of heuristic adversarial image transformations designed to degrade model performance while remaining human-interpretable. Through a two-phase evaluation of introductory assessments, we manually assess baseline VLM performance on circuit diagrams, followed by an automated large-scale evaluation of topological structures (logic gates) and coordinate geometry (Karnaugh maps). We find that while highly capable VLMs can exhibit appreciable robustness, all models suffer vulnerability to adversarial perturbations. We conclude that while visual perturbations act as a viable near-term stopgap, long-term assessment security requires educators to reapproach assessment design given continually increasing VLM performance.
[CV-6] ASTRA-SR: Atmospheric Seeing and Turbulence Restoration for Astronomical Image Super-Resolution ICASSP2027
链接: https://arxiv.org/abs/2609.26731
作者: Xining Ge,Ziteng Cui,Shuhong Liu
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: 4 pages of main text plus references, 4 figures. Submitted to ICASSP 2027
Abstract:Ground-based planetary imaging suffers from atmospheric turbulence, sensor noise, and limited sampling, making restoration a joint denoising, deblurring, and super-resolution problem. We present ASTRA-SR, a blind single-frame restoration framework trained on a physics-grounded synthetic dataset. High-dynamic-range spacecraft RAW observations serve as clean sources, and paired LR inputs are synthesized using measured layer-integrated turbulence strengths, propagated moving phase screens, exposure-averaged spatially varying PSFs, and sensor this http URL-SR first estimates a noise-suppressed but blur-retaining LR image, then restores spatial structure through multiscale processing and reconstructs HR detail with serial spatial-amplitude refinement. It yields a 0.49 dB foreground PSNR gain over the strongest baseline approaches.
[CV-7] GAD-MambaUNet: Direction-Group Mamba with Gradient-Adaptive DINOv3 Distillation for Lightweight Medical Image Segmentation
链接: https://arxiv.org/abs/2609.26729
作者: Fang Wang,Huitao Li,Wenhan Chao,Zheng Zhuo,Xinxin Yang
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:In this paper, we proposed GAD-MambaUNet, a lightweight medical image segmentation network that combines efficient local modeling, direction–group state-space interaction, and training-time foundation-model supervision. To improve contextual modeling in compact segmentation networks, we introduced Direction-Group Graph Selective Scan (DG-GSS), which treated scan-direction and channel-group responses as graph nodes and enabled structured information exchange before multi-directional fusion. We further incorporated DINOv3-GAD supervision, where a frozen DINOv3 teacher provided semantic guidance during training, and Gradient-Adaptive Distillation dynamically regulated the distillation strength. GAD-MambaUNet achieves a favorable accuracy–efficiency balance compared with representative lightweight and general segmentation methods. Ablation studies further verify the effectiveness of DG-GSS and training-time DINOv3-GAD supervision. In future work, we will explore more flexible teacher–student alignment strategies and extend the proposed framework to more diverse medical segmentation scenarios, such as multi-class and multi-modal segmentation tasks.
[CV-8] DIFTA-3D: Depth-Consistent Instance-Level Feature Transfer and Adaptation of DINOv3 for 3D Detection
链接: https://arxiv.org/abs/2609.26702
作者: Linman Wang,ZiFei Zhang,Chunran Zheng,Xiwang Dong,Jiarong Lin
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: 9 pages, 6 figures, conference paper
Abstract:RGB-D 3D instance detectors benefit from visual semantics, but the task-specific Faster R-CNN/ResNet branch used by IIFNet3D couples feature extraction to a separately trained 2D detector and its image-domain labels. Replacing that branch with a frozen vision foundation model removes this task-specific dependency, but may introduce occlusion noise and a mismatch between patch features and geometry-aware detection features. In this work, we investigate this replacement through an adaptation of DINOv3 to the instance-level fusion pipeline of IIFNet3D. At the core of our approach is a depth-consistent feature pipeline that projects scene points into calibrated RGB-D frames, applies a metric depth-residual check, averages the accepted DINOv3 features into an offline point cache, and aggregates the cached features inside proposal-aligned RoI grids. The geometric and bidirectional instance-fusion paths are preserved, while Conservative VAID is evaluated as a low-strength, support-weighted semantic distillation recipe applied only to positive RoIs. We conduct extensive evaluations on ScanNetV2 to assess the proposed transfer recipes. On ScanNetV2, our DINOv3 control achieves mAP scores of 76.15 and 60.93 at IoU thresholds of 0.25 and 0.50, respectively. The Conservative VAID setting achieves mAP scores of 76.59 and 62.16, corresponding to numerical gains of 0.44 and 1.23 points over the control, respectively, in this checkpoint-level recipe comparison. The reported IIFNet3D result of 75.7/63.8 is used only as an external reference because the visual branch and processing protocol differ. Accordingly, we interpret these results as evidence for a controlled transfer recipe rather than as a causal estimate of the individual contributions of VAID or depth filtering.
[CV-9] Longitudinal Retinal Vascular Remodeling in Myopic Children Treated with Orthokeratology or Defocus Lenses: A Two-Year Comparative Study
链接: https://arxiv.org/abs/2609.26662
作者: Zhihao Zhao,Yinzheng Zhao,Jie Zhang,Huiqin Jiang,Yanyu Shangguan,Yanfei Sun,Li Chen,Yanlong Bi,M. Ali Nasseri,Bing Li
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Purposes: To characterize longitudinal retinal vascular changes in myopic children treated with orthokeratology (OK) or multifocal defocus lenses (Defocus) and to examine their association with axial elongation. Methods: In this retrospective cohort study, 43 myopic children underwent comprehensive clinical examination and fundus photography at baseline, 12 months, and 24 months. Axial length (AL) and spherical equivalent refraction (SER) were recorded at baseline, 6, 12, and 24 months. An automated segmentation model extracted vascular parameters, main vessel angle (MA), branching angle (BA), bifurcation edge angle (BEA), crossover point (COP), and terminal vessel count (TVC). Repeated-measures ANOVA assessed temporal changes. Pearson or Spearman correlations evaluated associations between AL and vascular metrics. Results: Over 24 months, the OK group exhibited significantly slower axial elongation than the Defocus group (0.214 mm and 0.522 mm, p 0.01). In the OK group, MA and BA decreased modestly, BEA in arteries declined gradually, but COP and TVC remained relatively stable. The Defocus group demonstrated more pronounced decreases in MA and BA, an increase in BEA, and significant reductions in COP and TVC (p 0.05). Correlation analysis revealed stronger associations between AL and vascular parameters, especially COP and TVC, in the Defocus group at all time points, whereas only BA and BEA correlated with AL in the OK group. Conclusions: OK lenses mitigate axial elongation and induce milder retinal vascular remodeling compared to Defocus lenses. Distinct temporal patterns of vascular metrics changes were observed between the two interventions, and correlate differentially with axial growth.
[CV-10] ROAM-ASD: Robust Open-World Active Speaker Detection with Flexible Multimodal Fusion ICASSP2027
链接: https://arxiv.org/abs/2609.26648
作者: Pu Wang,Yujun Wang,Hugo Van hamme
类目: Multimedia (cs.MM); Computer Vision and Pattern Recognition (cs.CV); Sound (cs.SD); Audio and Speech Processing (eess.AS); Image and Video Processing (eess.IV)
备注: Submitted to IEEE ICASSP 2027
Abstract:Active speaker detection (ASD) requires reliable association between visible faces and acoustic speech, yet existing systems often degrade under challenging domains or incomplete observations. We introduce ROAM-ASD, a robust audiovisual framework that jointly models audio, full-face, and fine-grained mouth representations. A unified joint self-attention mechanism processes all input streams together with modality-agnostic query tokens, enabling direct interaction among available modality inputs. Modality dropout further improves robustness when input streams are unavailable. ROAM-ASD achieves state-of-the-art performance across five ASD benchmarks: 98.8% mAP on WASD, 87.9% on UniTalk, 96.5% on AVA, 99.3% on ASW, and 98.2% on Talkies, improving over previous best systems by 5.1, 4.7, 0.9, 1.0, and 2.1 mAP points, respectively. ROAM-ASD also substantially improves zero-shot cross-dataset generalization and remains robust to missing observations.
[CV-11] Laryngeal Structure Segmentation in High-Speed Videoendoscopy Using Deep Learning
链接: https://arxiv.org/abs/2609.26636
作者: Sardar Nafis Bin Ali,Mohsen Zayernouri,Dimitar D. Deliyski,Maryam Naghibolhosseini
类目: Computer Vision and Pattern Recognition (cs.CV); Image and Video Processing (eess.IV)
备注: 22 pages, 9 figures
Abstract:Laryngeal high-speed videoendoscopy (HSV) offers an effective means of observing the motion of different laryngeal structures along with vibratory behaviors of the vocal folds under various voicing conditions. Segmentation of laryngeal tissues enables analysis of different tissue structures and their dynamics, helping characterize the involvement of laryngeal muscles in voice production. Given the large number of HSV frames, automating this task is imperative. While deep learning-based methods have been implemented in previous studies to segment laryngeal structures, they have not been applied to HSV data during connected speech, which poses significant challenges due to excessive tissue movements and image quality limitations associated with fiberoptic image acquisition. The application of deep learning to connected speech data is critical for capturing nonstationary laryngeal behaviors and identifying anomalous patterns associated with voice disorders. The present study aims to address these gaps by training U-Net models to detect the aryepiglottic folds and arytenoid cartilages, vocal folds, epiglottis, and glottal area, using HSV data from both sustained vowel phonation and connected speech obtained from normophonic and disordered voices. Image pre-processing techniques, including noise removal and histogram equalization, were applied to improve the quality of the training HSV images and enhance network performance. Finally, to evaluate the accuracy and reliability of the networks, quantitative performance metrics were used alongside qualitative visual inspection of the test images. The high performance of the developed networks, with overall accuracies exceeding 95%, establishes their potential as reliable tools for automated laryngeal image analysis, quantitative characterization of laryngeal dynamics, and future detection of anomalous laryngeal behaviors in clinical settings.
[CV-12] Label-Efficient Learning for Ground-Based Sky-Image Classification: A Benchmark of Transfer Learning Active Learning and Pseudo-Labeling on GCD
链接: https://arxiv.org/abs/2609.26631
作者: Esther Bou Dagher,Viktoriya Bu-Dager,Boguslaw Zegarlinski
类目: Machine Learning (cs.LG); Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Accurate ground-based cloud classification is important for atmospheric monitoring, solar-energy forecasting, aviation weather assessment, and climate observation systems. However, reliable sky-image annotation is time-consuming, especially when cloud types are visually similar or mixed. We study the label efficiency of deep learning for ground-based cloud classification using the Ground-based Cloud Dataset (GCD). Rather than proposing a new architecture, we benchmark three practical strategies under limited annotation budgets: supervised transfer learning, uncertainty-based active learning, and high-confidence pseudo-labeling. An ImageNet-pretrained ResNet50 is used as a common frozen backbone, with experiments repeated over five random seeds for label budgets from 1% to 100% of the training labels. Supervised transfer learning is already highly label-efficient: test accuracy increases from 0.635 \pm 0.018 with 1% labels to 0.730 \pm 0.002 with 40% labels, approaching the full-label result of 0.735 \pm 0.003 . Active learning and pseudo-labeling are competitive with supervised sampling and provide small improvements for some metrics and budgets, but neither gives a large or consistent aggregate gain. Diagnostic analyses show that accepted pseudo-labels are reliable, with accuracy from 0.946 to 0.977 , but biased toward easier high-confidence sky-type groups. In contrast, uncertainty sampling preferentially queries visually challenging groups, including Mixed and the confusable Stratocumulus and Cumulonimbus groups, but these targeted acquisitions yield only modest gains. Overall, transfer learning substantially reduces annotation requirements for GCD, while simple active and semi-supervised strategies provide limited additional benefit over a strong supervised baseline.
[CV-13] A Data-Interventional Framework for Auditing Privacy and Fairness in Generative Medical Imaging
链接: https://arxiv.org/abs/2609.26623
作者: Mischa Dombrowski,Bernhard Kainz
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: Accepted for publication at the Journal of Machine Learning for Biomedical Imaging (MELBA) this https URL
Abstract:Diffusion-based synthetic data generation offers a promising route for sharing medical imaging data without releasing sensitive patient records. However, generative models face a fundamental tension between privacy and fairness: they may memorize rare training samples, leading to privacy risks, or fail to reproduce underrepresented features, resulting in unfair synthetic distributions. While prior work has largely focused on either memorization or fairness in isolation, their interaction remains insufficiently understood. In this work, we introduce a data-interventional framework to systematically analyze privacy and fairness in diffusion models. We discuss synthetic anatomical fingerprints (SAFs), rare and manually injected image features, as controlled probes to study whether models generalize sensitive attributes across identities, memorize training samples, or suppress rare signals entirely. Across multiple conditioning modalities, we observe a consistent behavior: models either forget these fingerprints or memorize the entire image in which they appear, but do not generalize them to novel images. To support large-scale auditing where explicit sample extraction is infeasible, we further introduce the indicator metric t’, which estimates a model’s susceptibility to memorization by exploiting the internal structure of the diffusion process. By comparing conditioning signals of varying surprisal, we reveal a clear relationship between conditioning rarity and memorization behavior. Highly surprising conditioning signals act as retrieval keys that amplify memorization, whereas low-surprisal conditioning signals systematically suppress rare features, even when these appear repeatedly in the training data. Our findings provide actionable insights and concrete mitigation strategies for safe and fair synthetic medical data sharing. Code is available at this https URL.
[CV-14] GeoComposer: Geometry-Grounded Photographic Composition Instruction
链接: https://arxiv.org/abs/2609.26620
作者: Shuangzhi Li,Qiaoqiao Jia,Xingxin Chen,Guile Wu,Dongfeng Bai
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Photographic composition aims to provide visual guidance for improving the framing, viewpoint, and spatial arrangement of an image. Early methods primarily rely on image cropping to enhance composition, which is restricted to the viewpoint and spatial arrangement of the input image. Recent methods have explored image understanding and editing to improve composition, but they mainly focus on instruction following and aesthetic quality, overlooking the importance of 3D scene geometry consistency for photographic composition. In this work, we propose GeoComposer, a novel geometry-grounded photographic composition framework that analyzes the composition of a given image to generate textual guidance and synthesizes a visual exemplar that enhances the composition of the given image. To promote geometry-grounded composition, we propose a geometry-aware representation learning mechanism that leverages geometric priors from a visual geometry foundation model to shape the intermediate representations of the composition editing model. This mechanism preserves both global structural relationships and local fine-grained correspondences for geometry-grounded composition. Furthermore, we propose a reinforcement learning strategy guided by a hybrid reward that jointly optimizes instruction following, aesthetic quality, and geometric consistency. This enables the model to generate visual exemplars that faithfully follow the composition instructions while remaining visually appealing and geometrically consistent. Extensive experiments show the superiority of our approach over state-of-the-art methods, highlighting its effectiveness in generating visually appealing and geometrically consistent composition.
[CV-15] MMAP: Multimodal Missing-Aware Pretraining for Longitudinal Alzheimers Prediction MICCAI
链接: https://arxiv.org/abs/2609.26617
作者: Fiona Kekwick,Matthew Baugh,Bernhard Kainz,Paul M. Matthews,Wenjia Bai
类目: Computer Vision and Pattern Recognition (cs.CV); Machine Learning (cs.LG)
备注: To be published in the proceedings of the 2026 MICCAI Workshop on Multimodal Learning with Medical Tabular Data
Abstract:Clinical decision making heavily relies on predicting the disease progression trajectory by seeking to understand patient’s health status which is characterised by multimodal medical data. AI holds great potential for learning useful representations from multimodal medical data to predict disease progression and aid clinical decision making. However, development of predictive AI models is constrained by missing modalities and incomplete tabular data frequently occurring in medical datasets. In addition, disease labels alone may only provide limited supervisory signals for learning representations from high-dimensional multimodal data. Here, we present MMAP, a novel Multimodal Missing-aware Alignment Pretraining method for learning image-tabular representations from incomplete data. An image encoder is pretrained with efficient sigmoid contrastive learning combined with generative reconstruction. A tabular encoder is built upon a tabular foundation model. A missing token generator enables the two encoders to take incomplete data as input, enabling the model to be robust against missing modalities, either with missing images or missing tabular data. We evaluate the clinical usefulness of the learnt multimodal representations on two challenging longitudinal clinical tasks for Alzheimer’s disease: predicting disease stage conversion and predicting amyloid status. The proposed method outperforms strong multimodal and unimodal baselines.
[CV-16] Foundation model embeddings capture pre-diagnostic changes on screening mammograms
链接: https://arxiv.org/abs/2609.26605
作者: Kalina P. Slavkova,Eric Brattain,Aditya Gowd,Akash Pattnaik,Jean-Benoit Delbrouck,Matthew Morgan,Julie Bauml,Javid Abderezaei,Khan Siddiqui
类目: Computer Vision and Pattern Recognition (cs.CV); Machine Learning (cs.LG)
备注: 13 pages, 5 figures, supplementary info attached
Abstract:Foundation model embeddings of screening mammograms may encode pre-diagnostic tissue change without task-specific adaptation. We tested whether embeddings move faster along a data-derived “cancer direction” in women later biopsied for cancer than in matched screen-negative controls, and whether this depends on pretraining domain. We studied 1,773 biopsied women (785 malignant, 988 biopsy-negative) and 1,773 matched controls, each with at least two annual screening exams before their index exam. An identical pipeline was applied to four 2D models: Mammo-CLIP (MC, out-of-distribution mammography), HOPPR (in-distribution mammography), MedImageInsight (MII, general medical imaging), and BiomedCLIP (biomedical vision-language pretraining on literature figures). Breast-level embeddings quantified longitudinal movement along the cancer direction. We compared cases and controls using a between-patient design with complementary mixed-effects analysis, and biopsied versus healthy contralateral breasts within patients. Under matched modality in MII embedding space, malignant cases drifted significantly faster than controls in the first two screening intervals preceding the index exam; biopsy-negative cases showed significance only in the first. MC differences were significant in the first interval for both biopsy groups. Within-patient comparisons showed a broadly similar pattern, with MC significance extending to the second interval in both groups and HOPPR showing significance at interval 1. BiomedCLIP showed no significant differences in either design or biopsy group. Overall, directional embedding velocity emerges as a property of clinically grounded rather than general biomedical pretraining, showing that foundation model embeddings can encode pre-diagnostic mammographic change without task-specific adaptation.
[CV-17] GTR: Gated Token Recurrence for Efficient Dense Prediction
链接: https://arxiv.org/abs/2609.26590
作者: Zhe Feng,Longfei Liu,Wei Liu,Kai Chen,Jiangjiang Kong,Wei Zhou,Yifeng Qian,Dexiong Chen,Xuanlong Yu,Xi Shen
类目: Computer Vision and Pattern Recognition (cs.CV); Machine Learning (cs.LG)
备注: Project page is available at: this https URL
Abstract:Self-attention-based vision backbones perform well on dense prediction, but the quadratic computational cost of global softmax attention limits their efficiency as image resolution increases. We introduce Gated Token Recurrence (GTR), a softmax-free recurrent vision backbone that combines gated linear attention, alternating spatial scan directions, and spatially enhanced SwiGLU blocks. GTR is distilled from a detection-specialized DINOv3 teacher using only final-layer patch-token alignment through a linear projection and squared \ell_2 loss, without masked-token prediction or intermediate-layer supervision. With Objects365 detector pre-training, GTR-L achieves 58.9 box AP on COCO \textttval2017 with 1.908,ms median batch-one latency under compiled FP16 execution on an RTX~4090. The same backbone also transfers to instance segmentation, pose estimation, oriented detection, semantic segmentation, and monocular depth estimation. In an isolated kernel benchmark, our specialized chunkwise CUDA operator is 4.0\times faster than FLA v0.5.0 at 1.6K tokens on RTX~4090. TensorRT deployment on DRIVE AGX Thor achieves 2.282–8.769,ms median batch-one latency across the evaluated models. These results show that recurrent token mixing can provide an efficient alternative to global softmax attention for high-resolution dense prediction and edge this http URL page: this https URL
[CV-18] Radiomics–Foundation Fusion for Interpretable RCC Classification: Internal Benchmarking and Exploratory External Transfer MICCAI2026
链接: https://arxiv.org/abs/2609.26578
作者: Yuan Liang,Fangyijie Wang,Kathleen M. Curran,Guénolé Silvestre,Sourav Bhattacharjee,Abraham Campbell
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: Accepted for an oral presentation at CaPTion 2026, a MICCAI 2026 workshop. 11 pages, 3 figures
Abstract:Accurate preoperative subtype classification of renal cell carcinoma (RCC) from contrast-enhanced CT remains clinically challenging because clear cell RCC (ccRCC) and non-clear cell RCC often show overlapping imaging appearances. This study evaluates whether foundation representations reduce reliance on handcrafted radiomics, or whether radiomics remains complementary for interpretable tumour characterisation. We compared radiomics, conventional CNN features, MedicalNet-pretrained features, MedVAE representations, and fusion variants for binary ccRCC classification on KiTS23, reporting area under the receiver operating characteristic curve (AUC) with bootstrap confidence intervals and average precision (AP) as a complementary class-imbalance-sensitive metric. We further assessed branch-removal ablation, TCGA/AIMI external transfer, and interpretability using radiomics permutation importance and gate-level analysis. Internally, 3D MedVAE gated fusion achieved the best performance, with an AUC of 82.7% and AP of 92.2%. On the external TCGA cohort, the same model achieved an AUC of 79.5% and AP of 98.9%, although specificity remains uncertain because only two external non-ccRCC cases were available. Gate analysis showed a radiomics-dominant fusion regime, suggesting that foundation representations acted as case-dependent refinement signals rather than replacements for structured tumour descriptors. These findings support radiomics as a complementary and clinically interpretable component of CT-based RCC characterisation in the foundation-model era.
[CV-19] Beyond End-Task Success: How to Audit Visual Experience Retrieval in Robotics IROS2026
链接: https://arxiv.org/abs/2609.26567
作者: Eshika Pathak,Leela Krishna
类目: Robotics (cs.RO); Computer Vision and Pattern Recognition (cs.CV)
备注: Accepted to the IROS 2026 Workshop on Embodied Neuro-Symbolic AI for Reliable and Safe Robotics (ReS AI)
Abstract:Robots that store past experiences must select which one to reuse in a new scene. Most systems select by visual similarity, and most evaluations report only the success of the selected experience. That number does not show whether the selection was good: a rule can score well by repeatedly using one broadly transferable experience, or poorly because its preferred experience is weak. Since robots increasingly adapt by reuse rather than retraining, a score that describes the library rather than the rule misleads what the field builds next. We contribute an audit methodology: execute every stored experience in every query scene, over two manipulation tasks, three reuse mechanisms, and libraries of K=3 , 10 , and 50 . Because every alternative’s outcome is known, a score can be traced to per-scene selection or to library quality. The audited rules select by nearest-neighbor distance in five visual embeddings, from raw pixels to CLIP. (1) One fixed experience, chosen with hindsight, captures 30-58% of the gap between random selection and an oracle; per-scene selection competes for the remaining 0.07-0.15 in success rate. (2) At K\ge10 , visual rules concentrate on one experience 1.5-3 times more than the oracle does, and their scores then follow that experience’s quality. (3) Wherever a rule differs significantly from a shuffle that keeps its selection rates but pairs them with scenes at random, the rule is worse, for every learned image policy. (4) Visual distance predicts well whether a given pair will succeed (AUROC up to 0.96), yet ranks the candidates within one scene no better than chance for four of five embeddings at K=50 (AUROC 0.45-0.52). Exhaustive execution is usually infeasible, so the audit reduces to two cheap reports any study can give: the distribution of selected experiences, and the success of the best single experience in hindsight.
[CV-20] Vision Foundation Models with Synthetic-Only Training for Monocular Spacecraft Pose Estimation IROS2026
链接: https://arxiv.org/abs/2609.26561
作者: John Church,Vazghen Nikolian
类目: Computer Vision and Pattern Recognition (cs.CV); Robotics (cs.RO)
备注: 6 pages, 3 figures, 4 tables. A shorter version was accepted to the IROS 2026 Space Robotics Workshop (non-archival)
Abstract:We present an improvement on previous spacecraft pose estimation architectures that results in the lowest published mean rotation errors we know of on the SPEED+ lightbox and sunlamp test sets for a known, non-cooperative spacecraft. By using a previously established heatmap-based pose estimation architecture and adapting a large self-supervised ViT foundation model (DINOv3) in place of the smaller convolutional and ViT encoders of previous work, we show that pose estimation accuracy improves from 300M to 840M parameters with no saturation yet observed. We also evaluate our 840M model on a Jetson Orin NX 16GB, measuring single-pass network inference at 133.8 ms per crop with a board draw of 32.0 W. These measurements demonstrate embedded inference feasibility on a processor family with orbital flight heritage. Our resulting model outperforms previous models across lightbox and sunlamp domains while training only on synthetic data. Our best model, using DINOv3 840M adapted with LoRA as the encoder (rank 64, three-seed ensemble with four-rotation test-time augmentation), results in 1.56^\circ mean rotation error on sunlamp and 1.17^\circ on lightbox, compared to the previous best mean rotation errors we know of on these test sets, 2.66^\circ and 1.75^\circ by EagerNet.
[CV-21] Latent Commonality Expectation-Maximisation for Box-supervised Tree Crown Instance Segmentation
链接: https://arxiv.org/abs/2609.26549
作者: Thomas Pitts,Kunqi Li,Bin Liang
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: 37 pages, 18 tables. Code and model checkpoints to be released upon publication
Abstract:Individual tree crown segmentation from aerial imagery underpins tree-level carbon accounting, biodiversity, and restoration monitoring at landscape scale. However, existing models are predominantly trained on dense canopy forest imagery and degrade in savannah and drylands, where tree crowns are sparse, of variable appearance, and underrepresented in annotated benchmarks. These models also typically depend on costly polygon annotations. We introduce LACE (LAtent Commonality Expectation-maximisation), a box-supervised instance segmentation model, evaluated on 0.1 m/px aerial RGB tree crown imagery. LACE uses a frozen DINOv3-web ViT-L/16 encoder, applied at four spatial offsets and interlaced into a denser feature grid, with a lightweight CenterNet-style detection head trained solely on bounding boxes. We use expectation-maximisation to separate recurring appearance, the “treeness”, within bounding boxes from surroundings. On the OAM-TCD benchmark test set, LACE reaches a mask AP _50 of 0.663 \pm 0.001 (3 seeds) trained on 900 box-annotated images and without mask annotations, above the 0.626 scored by Restor’s released mask-supervised Mask R-CNN, which was trained on the full ~4.2k image set. On a sparse-canopy holdout set, mask AP _50 rises to 0.691 versus 0.612 for Detectree2, a mask-supervised baseline. On NeonTreeEvaluation, using the official evaluation code, LACE reaches 0.728 \pm 0.003 F1@0.4 (5 seeds) from 23,424 hand-annotated RGB boxes alone, matching the authors’ DeepForest model’s published 0.719, using under 0.1% of its training annotations and none of its LiDAR-derived 30M-crown pretraining set. By leveraging frozen self-supervised features, LACE matches or surpasses fully-supervised specialist baselines from boxes alone, removing the need for polygon annotation in tree crown instance segmentation for sparse-canopy environments where labelled data is scarce.
[CV-22] Notes on Fourier-Bessel wavelets
链接: https://arxiv.org/abs/2609.26537
作者: Marcel Venturotti,Georgios Exarchakis
类目: Machine Learning (cs.LG); Computer Vision and Pattern Recognition (cs.CV); Numerical Analysis (math.NA)
备注:
Abstract:These notes develop the mathematical foundations and construction of a Fourier-Bessel wavelet family inspired by the disk harmonics of Shaqfa et al.[9]. We begin with the relevant properties of Bessel and modified Bessel functions and introduce the wavelet properties required for the construction. We then derive the Fourier-Bessel disk harmonics as solutions to the Helmholtz equation on the unit disk subject to a Neumann boundary condition. Building on this basis, we construct a wavelet family by applying a Gaussian spatial envelope and introducing a zero-mean correction for the zeroth angular order. We derive the corresponding normalisation constants for L^2 -based applications and discuss L^1 -based normalisation for frequency-domain peak consistency. Finally, we derive a closed-form Fourier-domain representation of the resulting wavelets. The main motivation is the approximately linear spacing, which converges to \pi between consecutive radial eigenvalues. Rather than replacing the conventional dyadic organisation of wavelet families, this construction lays out the foundation to explore whether a more uniform radial frequency allocation can be useful for applications in which broad and balanced frequency coverage is desirable. Subjects: Machine Learning (cs.LG); Computer Vision and Pattern Recognition (cs.CV); Numerical Analysis (math.NA) Cite as: arXiv:2609.26537 [cs.LG] (or arXiv:2609.26537v1 [cs.LG] for this version) https://doi.org/10.48550/arXiv.2609.26537 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
[CV-23] Virtual Encoders in Multimodal Transformers
链接: https://arxiv.org/abs/2609.26513
作者: Katsuya Ogata,Yuta Nakashima
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Multimodal language models traditionally rely on dedicated perceptual encoders to construct task-usable representations. More integrated architectures have recently emerged, which instead expose the shared transformer to lightly projected patches, audio frames, or discrete visual tokens. Where does this encoding happen when such representations are not provided? We find that the transformer can internalize this missing computation, constructing task-usable perceptual representations within its own early-to-middle layers before the downstream language model. We call this computational structure a Virtual Encoder. Across linear probing, similarities to perceptual encoders, and causal analyses, we identify signatures of this structure in models that receive perceptual tokens without continuous encoder-derived features. These analyses also suggest that the boundary between perception and language processing need not coincide within an architectural module. Instead, encoder-like computation can emerge as a functional regime within a shared transformer, providing a new perspective for understanding where and how multimodal models process perception.
[CV-24] Semantically-Guided Domain Randomization for Industrial Object Detection in Low-Image-Budget Regimes
链接: https://arxiv.org/abs/2609.26505
作者: Jose Moises Araya-Martinez,Gautham Mohan,Jens Lambrecht
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Retraining visual perception pipelines in High-Mix, Low-Volume (HMLV) automotive manufacturing must be carried out under tight annotation, energy, and time budgets, yet most Synthetic Data Generation (SDG) strategies still operate in the thousands of images. This work evaluates Semantically-Guided Domain Randomization (S-GDR), an annotation-free adaptation pipeline that couples Vision-Language Model (VLM)-based semantic captioning of a small unannotated real reference set with diffusion-based background synthesis (Stable Diffusion XL (SDXL) conditioned by ControlNet and IP-Adapter) and mask-based object composition. On an automotive multi-object detection benchmark and with a fixed budget of 200 synthetic training images, S-GDR reaches mAP50-95 = 0.739 on a real held-out test set, outperforming a domain-randomized render baseline (mAP50-95 = 0.697) as well as brightness filtering, perceptual hashing, CycleGAN style transfer, and unguided diffusion variants sharing the same 200-image budget. These initial observations position S-GDR as a promising annotation- free alternative for extreme data-scarcity regimes.
[CV-25] Radiomics-Conditioned Modulation of RenalCLIP Features for Clear Cell Renal Cell Carcinoma Classification
链接: https://arxiv.org/abs/2609.26492
作者: Yuan Liang,Sourav Bhattacharjee,Abraham Campbell
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
备注: Accepted at the 7th International Conference on Medical Imaging and Computer-Aided Diagnosis (MICAD 2026). 10 pages, 2 figures
Abstract:Radiomics provides quantitative descriptions of tumour appearance that may complement disease-specific foundation models in small labelled cohorts. We investigate this complementarity for computed tomography-based classification of clear cell renal cell carcinoma. Our framework uses radiomics to modulate RenalCLIP features through feature-wise linear modulation (FiLM), while retaining a direct radiomics contribution. Internal testing and external validation compare it with conventional fusion strategies and reference classifiers. The FiLM model achieves an area under the receiver operating characteristic curve (AUC) of 0.804 internally and 0.854 externally, with the highest mean AUC among the evaluated RenalCLIP fusion strategies in both cohorts. Pathway ablations examine the contributions of conditional modulation and the direct radiomics residual, while feature permutation highlights the role of tumour texture. These findings support radiomics as a useful complement to RenalCLIP in a small labelled cohort and identify FiLM as an effective approach to integrating their representations for robust renal tumour classification.
[CV-26] From Token Importance to Conditional Removability: Rethinking Visual Token Pruning in Multimodal Large Language Models
链接: https://arxiv.org/abs/2609.26484
作者: Shengli He,Yongchao Liang,Roumeng He,Junjie Zeng,Jiyuan He,Xin Fang,Can Wu,Li Zheng
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Training-free visual-token pruning often uses token importance, redundancy, or related selection criteria as proxies for safe removal. We show that these signals alone do not fully characterize removability, which is conditioned on both representation depth and the surrounding deletion set. Controlled interventions demonstrate that removing the same tokens at different depths produces substantially different downstream perturbations, while changing only the deletion context at a fixed depth alters candidate marginals and pruning-boundary decisions. These findings show that token importance alone cannot determine when a token is safely removable or how its removability changes under joint deletion. Motivated by this perspective, we propose CoRePrune, a training-free two-stage framework. Progressive Perturbation-Aware Visual Pruning refreshes deletion effects as visual representations evolve, while Set-Conditioned Refinement reevaluates candidate rescue benefits under the current deletion set after visual–text interaction. Across five multimodal large language model backbones covering standard images, high-resolution inputs, and video, CoRePrune preserves performance under aggressive token budgets. On Qwen3.5, with a final budget of 128 visual tokens, it retains 90.3% of dense-model performance while reducing aggregate prefill time by 51.0%.
[CV-27] PP-Net: A Hybrid Physical-Prior Neural Network for Scattered Light Removal in Biomedical Images on Embedded Devices
链接: https://arxiv.org/abs/2609.26474
作者: Yongfei Guo,Tingjin Chu,Mengzhuo Liu,Hongwei Lou,Yuanhao Gong
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI); Machine Learning (cs.LG); Image and Video Processing (eess.IV); Signal Processing (eess.SP)
备注:
Abstract:Scattered light is common in biomedical images, yet its removal remains challenging. The difficulty arises from three aspects: first, aligned scattered-light-free biomedical ground truth is often unavailable; second, scattering is coupled with weak illumination and sensor-induced noise; and third, many learning-based restoration models are computationally expensive for embedded devices in Internet of Medical Things (IoMT) scenarios. To address these issues, this paper proposes PP-Net, a hybrid physical-prior neural network for biomedical scattered light removal. The proposed method consists of three components: DFN-Net suppresses sensor-induced noise, ASAP estimates the scattering map and recovers a physics-based prior map, and GF-Net refines the prior map by fusing it with the denoised observation. To reduce the dependence on paired biomedical ground truth, a progressive synthetic training and cross-domain transfer strategy is developed. Experiments show that the physical-prior branch improves the peak signal-to-noise ratio (PSNR) by up to 1.26 dB on paired synthetic benchmarks. Under joint noise-and-scattering degradation, PP-Net improves PSNR by more than 10.8 dB and the structural similarity index measure (SSIM) by more than 0.62 compared with representative baseline methods. On real W2S biomedical images, the proposed method reduces the average Natural Image Quality Evaluator (NIQE) score by 43.3%. Edge deployment with RKNN conversion and INT8 quantization achieves an average inference latency of approximately 200 ms per 512\times512 image over 360 test images. These results demonstrate that PP-Net provides an effective and deployable solution for microscopic imaging, endoscopic inspection, and edge-assisted biomedical analysis in IoMT scenarios.
[CV-28] Complementary Roles of Radiomics and Foundation Representations in Renal Cell Carcinoma Classification: A Comparative Study of 2D and 3D CT Encodings
链接: https://arxiv.org/abs/2609.26463
作者: Yuan Liang,Sourav Bhattacharjee,Abraham Campbell
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
备注: Accepted at Medical Image Understanding and Analysis (MIUA 2026). 15 pages, 2 figures
Abstract:Accurate preoperative subtype classification of renal cell carcinoma (RCC) from contrast-enhanced computed tomography remains clinically challenging. Radiomics provides structured tumour descriptors, whereas foundation representations offer transferable image features. However, it remains unclear whether radiomics still adds value beyond pretrained representations, and how 2D and 3D MedVAE encoders compare in this setting. We compared handcrafted radiomics, 2D MedVAE, 3D MedVAE, and their fusion for binary clear-cell RCC versus non-clear-cell RCC classification on KiTS23 under a unified preprocessing pipeline. Concatenation, cross-attention, and gated fusion were evaluated as representative integration strategies, and radiomics feature importance was analysed to support decision-centric interpretability. Fusion consistently improved discrimination over image-only MedVAE branches. The best overall performance was achieved by 3D gated fusion, with an AUC of 82.7%, outperforming the best 2D fusion model (79.6%), the radiomics baseline (74.4%), and the single-modality MedVAE branches. Ablation analysis further showed clear gains of the full fusion model over both image-only and radiomics-only variants, indicating complementary contributions from radiomics and image representations. These findings suggest that radiomics remains relevant for RCC CT classification in the presence of foundation representations, and that its integration with MedVAE is more effective in the 3D setting. More broadly, the study supports a complementary role for radiomics and foundation representations in clinically meaningful imaging decision support. Comments: Accepted at Medical Image Understanding and Analysis (MIUA 2026). 15 pages, 2 figures Subjects: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI) Cite as: arXiv:2609.26463 [cs.CV] (or arXiv:2609.26463v1 [cs.CV] for this version) https://doi.org/10.48550/arXiv.2609.26463 Focus to learn more arXiv-issued DOI via DataCite (pending registration) Submission history From: Yuan Liang [view email] [v1] Tue, 22 Sep 2026 14:14:33 UTC (352 KB)
[CV-29] Code Plans Diffusion Renders: Open-Ended Generative World Modeling
链接: https://arxiv.org/abs/2609.26458
作者: Zixun Fang,Yawen Shao,Kai Zhu,Jie Xiao,Shihan Chen,Yu Liu,Xueyang Fu,Yang Cao,Wei Zhai,Zheng-Jun Zha
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: this https URL
Abstract:We introduce \textbfCoDeR, a new paradigm for world modeling. Unlike existing video world models that implicitly represent world dynamics through visual observations, our system explicitly constructs an executable world with code and employs video generation models for visual realization. Specifically, we coordinate five complementary roles to translate high-level concepts into structured world rules, executable dynamics, and perceptual observations. This design enables \textitlong-term memory, \textitopen-ended interactions, \textitautonomous world evolution, and \textitmulti-agent scenarios, where multiple entities can act, interact, and evolve persistently beyond the current observation. Extensive experiments demonstrate that our framework substantially extends the capabilities of existing world models, enabling long-term memory, open-ended interactions, autonomous evolution, and persistent multi-agent dynamics, while achieving state-of-the-art performance across multiple evaluation settings. Code and model weights will be made publicly available. Project Page: \hrefthis https URLCoDeR.
[CV-30] Mammo-LIFE: Longitudinal Mammographic Imaging and Clinical Feature Enrichment for Post-Radiotherapy Outcome Prediction
链接: https://arxiv.org/abs/2609.26443
作者: Farnoush Bayatmakou,Maryam Hosseini,Reza Taleei,Arash Mohammadi
类目: Computer Vision and Pattern Recognition (cs.CV); Image and Video Processing (eess.IV)
备注:
Abstract:Recent advances in Artificial Intelligence (AI)-powered Computer-Aided Diagnosis (CAD) systems have substantially improved breast cancer screening, diagnosis, and prognosis. Comparatively, postradiotherapy outcome prediction using paired longitudinal mammograms has received considerably less attention. This is largely due to the limited availability of well-annotated longitudinal datasets. Longitudinal mammograms, coupled with paired pre- and post-treatment information, provide a unique opportunity to characterize treatment-induced breast tissue changes following radiotherapy. The resulting learned representations can serve as a valuable asset for advancing personalized radiotherapy planning and post-treatment management. In this context, we propose Mammo-LIFE, a patient-level multimodal framework for post-radiotherapy outcome prediction that combines longitudinal mammographic features with patient-level clinical variables. The imaging branch processes paired pre- and post-treatment mammograms acquired from the four standard views using a mammography-specific encoder adapted via Low-Rank Adaptation (LoRA). Within each view, preand post-treatment representations are explicitly compared through a longitudinal comparison module to capture treatment-related changes. The resulting view-level embeddings are then aggregated using learned view-attention pooling to form a unified patient-level mammographic representation. Selected clinical variables are subsequently combined with the image-derived prediction probability through a late-fusion strategy. To evaluate the effectiveness of combining paired longitudinal mammograms with clinical information, experiments were conducted on an in-house clinical cohort using patient-level stratified five-fold cross-validation.
[CV-31] Latent Dataset Distillation for Human Motion Prediction
链接: https://arxiv.org/abs/2609.26430
作者: Ge Tian,Guang Li,Takahiro Ogawa,Miki Haseyama
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Dataset distillation (DD) compresses a large training set into a compact synthetic set while preserving downstream training utility. Although DD has been widely studied for images and recently extended to time-series forecasting, its application to human motion prediction remains largely unexplored. Human motion is high-dimensional and structurally coupled, and gradient matching (GM) in the original motion space optimizes many correlated variables without a prior on pose plausibility or temporal dynamics, which frequently yields implausible and unstable synthetic motions. To address this limitation, we propose a latent DD framework that regularizes distillation with a learned motion prior. Motions are first compressed by a residual-quantized variational autoencoder (RVQ-VAE), and distillation then updates only a learnable latent bank through the frozen quantizer and decoder. The pretrained decoder restricts synthetic motions to its output space, while residual quantization progressively refines the latent approximation across multiple codebooks and alleviates the representational bottleneck of single-stage vector quantization. Experiments on Human3.6M, CMU, and 3DPW with two prediction backbones show that the proposed framework outperforms direct GM in 27 of 30 evaluated settings and random subsets in every setting, and produces visibly more plausible synthetic motions in qualitative comparisons.
[CV-32] QuantWM: Temporally Consistent 2-Bit KV Cache Quantization for World Models and Video Generation
链接: https://arxiv.org/abs/2609.26425
作者: Jiaqi Zhao,Xiaobin Hu,Bo Yin,Junpeng Jiang,Miao Zhang,Shuicheng Yan
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
备注:
Abstract:KV cache memory has become a major deployment bottleneck for video generation and world models, which motivates low-bit quantization study for efficiency. Existing 2-bit KV cache quantization methods can achieve nearly lossless performance on video benchmarks such as VBench, however, we find that they still cause severe temporal flickering and visual degradation. Meanwhile, deeper investigates show that Key quantization produces smaller reconstruction errors than Value, but surprisingly leads to much larger output degradation. We trace this discrepancy to attention: small Key perturbations can change the attention logits, i.e., QK^\top, and shift the temporal-spatial tokens selected by Queries. These observations motivate us to explicitly preserve attention logits and temporal-spatial token selection during KV cache quantization to alleviate the visual degradation problem. To address this issue, we present QuantWM, a training-free and strictly causal 2-bit KV cache quantization framework. QuantWM introduces two complementary techniques to mitigate the attention shifts. Firstly, quantization-sensitivity-aware clustering (QSAC) jointly considers historical Query sensitivity and residual ranges to select INT2-friendly Key centroids, which reduces quantization errors in channels that are more critical to attention. In addition, principal-subspace attention compensation (PSAC) restores the remaining Key errors along the dominant Query subspace using low-rank projections, which provides a direct and efficient correction to stabilize attention logits. Extensive experiments on Causal-Forcing, LingBot-World-v2, HY-World 1.5, Matrix-Game-2 and Longcat-Video demonstrate that QuantWM significantly improves visual quality and temporal consistency, while outperforming existing methods across image and video quality metrics with up to 6.20x KV cache memory compression and limited additional overhead.
[CV-33] Sample Simulate Select: Physics-in-the-Loop Text-to-Motion for Humanoids Without Training
链接: https://arxiv.org/abs/2609.26420
作者: Raphael Memmesheimer,Sven Behnke
类目: Robotics (cs.RO); Computer Vision and Pattern Recognition (cs.CV)
备注: 8 pages, 9 figures, 5 tables
Abstract:Text-to-motion models generate plausible human motion but do not model a robot’s dynamics; whole-body tracking controllers execute robot references reliably but cannot replan an infeasible one. Recent language-to-humanoid systems bridge this gap by training. We measure how much of the gap closes with no training at all, by putting the deployment controller itself in the loop. Sample-simulate-select (S ^3 ) draws N motions per prompt from a frozen text-to-motion model, retargets each to a Unitree G1 by direction-matching inverse kinematics, rolls all of them out under full rigid-body dynamics with the pretrained SONIC tracking policy, and keeps the candidate the policy executed best. Because the verifier is the deterministic simulator itself, S ^3 attains the any-of- N ceiling by construction; what we measure is where that ceiling lies and what falls short of it. On 200 stratified HumanML3D test prompts with N=8 , upright execution rises from 83.5% to 89.5% and hardware-gate passes from 33 to 85; on the complete test split (4,184 prompts) it rises from 80.5% to 89.5%. A kinematic verifier that predicts falls well (AUROC 0.90) recovers only a quarter of this gain: ranking a prompt’s own candidates is harder than classifying the population. What selection cannot fix is one class, prompts that lower the pelvis, which a generator trained on retargeted robot data does execute. We further score the semantic fidelity of the executed motion with the standard text-motion evaluator, with a real-mocap control that attributes the loss to the robot projection, ablate the retargeter against GMR (complementary failures: the any-of-8 ceiling rises to 95.0% over both), and execute all 177 gate-selected clips on the real G1: every one completes standing, with hardware tracking error matching simulation ( r=0.94 ).
[CV-34] MAVP: Map-Aware Visuomotor Policies for Mobile Manipulation
链接: https://arxiv.org/abs/2609.26378
作者: Jinhe Tang,Ruixiao Dai,Weiming Zhi
类目: Robotics (cs.RO); Artificial Intelligence (cs.AI); Computer Vision and Pattern Recognition (cs.CV); Machine Learning (cs.LG)
备注:
Abstract:Successful mobile manipulation requires coordinated base and arm motion while maintaining accurate spatial positioning. However, demonstration-trained policies can struggle to realise the intended base motion reliably, leading to spatial misalignment and subsequent manipulation failures. We present MAVP (Map-Aware Visuomotor Policies), a framework that improves execution reliability by predicting explicit base-pose targets and tracking them using localisation feedback. MAVP reconstructs a static map from teleoperated demonstrations and expresses demonstrated base trajectories in a shared map frame, providing consistent spatial supervision across demonstrations. At execution time, the policy receives RGB observations, joint states, and the robot’s current map-frame base pose, and jointly predicts target base poses, arm actions, and gripper actions. A low-level controller tracks the predicted base targets using feedforward motion and pose error feedback, enabling correction of execution deviations. We additionally use pose-noise augmentation during training to improve robustness to errors in the policy’s pose input. Across six real-world manipulation tasks and three policy families, MAVP achieves higher task success rates than unanchored velocity control in all tasks. Videos and additional results are available at this https URL.
[CV-35] KwaiMind Technical Report UAI
链接: https://arxiv.org/abs/2609.26375
作者: Junlong Wu,Zijun Li,Yuting Hu,Jia Sun,Pengcheng Wei,Yimin Zhou,Honglie Wang,Huaiqing Wang,Dewen Fan,Fei Zuo,Haixuan Gao,Lihui Peng,Tingxuan She,Yuqing Li,Boheng Zhang,Fan Yang,Wenwu Ou
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: KwaiMind Team, Kuaishou Group
Abstract:Commercial image editing requires product identity preservation, accurate text rendering, and user appeal alongside general editing quality. We present KwaiMind, an image editing system combining general capabilities with e-commerce specialization. An agent-based data engine maintains approximately 1.8 million high-quality editing pairs. Built on a multimodal diffusion transformer, KwaiMind undergoes continued pre-training and supervised fine-tuning, followed by preference optimization and online reinforcement learning. A general-purpose vision-language judge and specialized rewards for click-through rate (CTR), text rendering, and product consistency guide specialized policies, which are consolidated through on-policy distillation. We introduce Ecom-Bench, covering 11 commercial editing tasks with task-specific visual evaluation and CTR-based ranking. KwaiMind achieves the strongest overall scores among evaluated open-source editors on ImgEdit, GEdit, both language splits of REDEdit, and Ecom-Bench visual quality, and the highest aggregate CTR ranking score among compared systems. Offline, CTR-guided optimization increases the proportion of generated images whose predicted CTR exceeds that of the original product image from 12.16% to 37.41%. In an online A/B experiment, CTR-based selection of product main images yields an approximately 2.44% relative increase in actual CTR. These results demonstrate the value of domain-specific data and reward-driven alignment for commercial image editing.
[CV-36] On the Role of the Projector in Contrastive Self-Supervised Learning: Last-Layer Rank Dynamics Drive Representation Quality
链接: https://arxiv.org/abs/2609.26334
作者: Siladittya Manna,Priyangshu Mandal,Umapada Pal,Saumik Bhattacharya
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: Under review at Transactions on Machine Learning Research (TMLR)
Abstract:The dimensional collapse of representations in self-supervised contrastive learning is an ever-present issue. One notable technique to prevent such a collapse of representations is using a multi-layered perceptron network called Projector. In several works, the projector has been found to heavily influence the quality of representations learned in a self-supervised contrastive pre-training task. However, the question still lingers. What role does the projector play? Assuming the projector mitigates dimensional collapse, what prevents the terminal layer of the base encoder from functioning as the projector in the absence of an explicit multi-layer perceptron (MLP) head? In this work, we intend to study what happens inside the projector by examining the rank dynamics of the same and the encoder through empirical study and analysis. Through mathematical analysis, we observe that the effect of rank reduction predominantly occurs in the last layer. Motivated by this insight, we propose a weight regularization strategy applied specifically to the last layer. We demonstrate that this targeted approach yields better performance than applying orthogonal weight regularization across the entire network (WeRank), both with and without a projector. Our method improves Top-1 accuracy by more than 1% on SimCLR on the ImageNet100 dataset and consistently outperforms baseline SimCLR variants on CIFAR datasets, supporting our interpretation of the projector’s role.
[CV-37] Leverag ing Vision-Based Point Cloud Map Priors for Camera-Based 3D Object Detection and Online Vectorized HD Mapping IROS2026
链接: https://arxiv.org/abs/2609.26325
作者: Markus Käppeler,Rohit Mohan,Abhinav Valada
类目: Computer Vision and Pattern Recognition (cs.CV); Robotics (cs.RO)
备注: IROS 2026 Workshop on Long-Term Perception for Human-Centric Autonomy
Abstract:Camera-based 3D object detection and online vectorized HD mapping provide compact scene representations for autonomous driving, but both depend on accurate metric geometry and remain limited by depth ambiguity. Over long-term deployment, observations from repeated traversals can be accumulated into persistent point cloud priors that provide geometric context beyond the current observations. Existing explicit point cloud prior approaches, however, rely on LiDAR-based map construction and therefore require expensive 3D ranging sensors. We propose a framework that constructs a static point cloud prior map from previous camera traversals using Pi3X and augments each point with DINOv3 features. At runtime, a local prior patch is retrieved using global localization, encoded with a sparse voxel backbone, and fused in bird’s-eye view (BEV) with lifted multi-view camera features. Task-specific sparse transformer heads then predict 3D objects and vectorized map elements from the fused representation. On Argoverse 2, the vision-based prior improves a strong baseline from 0.287 to 0.299 CDS and from 0.669 to 0.750 vectorized mapping mAP. Ablations show that semantic DINOv3 features are particularly important for vectorized mapping. These results demonstrate that vision-built geometric-semantic priors provide an effective form of long-term scene memory for camera-based perception, improving both tasks without LiDAR for prior-map construction or online inference.
[CV-38] ForeDrive: Foresight-Guided End-to-End Autonomous Driving with a Planning -Relevant Latent World Model
链接: https://arxiv.org/abs/2609.26299
作者: Sinuo Wang,Zichong Gu,Yuhan Huang,Wenxin Wen,Xun Yang,Yiqing Zhang,Xingyu Zhang,Ningyu Che,Jie Ling,Qiankun Yu,Wei Liu,Jing Xu,Xinggang Wang
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: 9 pages, 4 figures; 8 pages supplementary with 4 figures
Abstract:Existing latent world models are typically optimized for future predictability, yet the resulting representations are not necessarily useful for planning in autonomous driving. Predictions are commonly used for pretraining or auxiliary supervision rather than as direct conditioning signals for trajectory generation. We propose ForeDrive, which learns a planning-relevant latent representation and couples it asymmetrically to a Diffusion Transformer (DiT) planner. The planner consumes multi-horizon latent future representations learned with a JEPA-style world model; planning gradients update the shared online encoder, while stop-gradient routing trains the latent predictor with forecasting losses only. Because predicted futures have varying reliability across horizons and BEV trajectories are misaligned with image tokens, we use gated visual fusion, future-status injection, and Trajectory-Adaptive Bias (TAB) to inject future latents as guidance without overriding the current observation. Trained with pure imitation learning and using only the current front-view image as visual input at inference, ForeDrive attains 89.9 PDMS on NAVSIM v1 and 90.0 one-stage EPDMS on NAVSIM v2, without reinforcement learning or an external trajectory scorer.
[CV-39] AIGC Video Detection based on the fusion of spatial-frequency-optical flow multimodal features
链接: https://arxiv.org/abs/2609.26274
作者: S. Hong,X.Q. Wang,C. Zhang,J.C. Wang,P.X. Duan,Y.W. Wang
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
备注:
Abstract:The rapid evolution of generative AI (e.g., Sora, Hunyuan) makes it essential to develop effective detection strategies that can generalize across ever-evolving synthesis techniques. This study is motivated by the observation of a fundamental challenge in generative models: the inherent difficulty of maintaining cross-modal consistency between appearance and motion. To this end, we propose a multi-modal framework for AIGC video forgery detection tasks, named Cross-Attention based Video Forgery Detector (CrossAtt-VFD), based on joint multi-view analysis of this http URL, we introduce a dual-branch architecture that simultaneously extracts spatial-frequency and optical-flow this http URL approach enables the modeling of videos from complementary perceptual this http URL core of this process is a dedicated cross-attention mechanism, which governs the alignment of the two modalities and translates cross-modal inconsistencies into a potent diagnostic signal. This multi-modal strategy facilitates the detection of motion that is statistically inconsistent with the visual appearance of a scene. Comprehensive experimental results demonstrated that our model achieves an accuracy of 94.22%, a precision of 91.67 %,and a recall of 96.25 %, effectively verifying the advantages of the multi-modal fusion strategy.
[CV-40] COVER: Codec-Robust Video Watermarking with Generative Video Priors
链接: https://arxiv.org/abs/2609.26236
作者: Yuxin Cao,Hao Yang,Ziqi Ding,Jie Hao,Wei Song
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: 10 pages, 5 figures
Abstract:Video watermarking underpins copyright protection and provenance for generated media, yet almost every video is compressed by a codec before it is stored or shared. A codec discards precisely the perceptually redundant components that most watermarks rely on, so the payload is often lost even when the marked video looked flawless beforehand. Existing methods leave this path open, since they treat compression as one entry in a generic list of distortions, while a real codec is not differentiable and cannot enter gradient-based training. We present COVER, the first learned video watermark built around codec compression as its design target, which survives that compression by embedding the payload in the latent space of a frozen generative video autoencoder and recovering it by re-encoding the received video into that same latent space. To make codec robustness trainable, we build a differentiable codec surrogate bank that simulates the dominant degradation modes of practical compression, and we train the embedder and the latent decoder through three shared recovery paths under a fidelity objective that constrains the residual in the pixel and frequency domains. Across four codecs at 12 settings, COVER attains 93.72% average bit accuracy, ranks first on 11 of the 12, improves the strongest prior method by 2.68 points, and lifts the worst operating point from 68.90% to 73.72% while each marked video stays visually close to the source clip that produced it.
[CV-41] he Temporal Moderation Gap: Text-to-Video Safety Filters Are Blind to Harm in Motion
链接: https://arxiv.org/abs/2609.26233
作者: Yuxin Cao,Fusen Guo,Yuezhong Wu,Huadong Mo,Wei Song
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: 12 pages, 2 figures
Abstract:Text-to-video (T2V) services inherit their safety stack from image generation, pairing a keyword prompt filter with a per-frame checker that blocks a clip whenever one sampled frame looks unsafe. This stack has a blind spot unique to video. We prove that any moderator ignoring frame order accepts a harmful clip whenever it accepts that clip’s benign shuffle, so harm carried by the ordering alone escapes. Empirically, the unmodified benchmark prompt already lands a clip in this moderation gap on 32.7% of Sequential-Action targets over four held-out seeds, and paraphrasing, scene splitting, and a feedback-driven prompt search show no significant improvement (paired McNemar p\ge0.12 ), so prompt engineering is not needed to expose the vulnerability. Dense-scoring all 97 rendered frames shows that about a third of the delivered clips merely hide an unsafe frame, while the rest stay harmful as ordered videos even though every frame passes, an order-blind residual the unmodified prompt reaches on a quarter of Sequential-Action targets. We also document a measurement pitfall, since scoring a searched prompt on its own render seed inflates a 7.5% per-generation rate into an apparent 46.7%. A user study confirms that people read these clips as harmful and their shuffles as safe. The fix is to read frame order, and an order-aware detector separates these clips from their own shuffles at AUC 0.74 where per-frame checking sits at chance, which is the signal deployed moderation throws away.
[CV-42] LLaVA-Assessor: Building the Foundation LMM For Visual Quality Assessment
链接: https://arxiv.org/abs/2609.26205
作者: Ziheng Jia,Zicheng Zhang,Jiaying Qian,Guangtao Zhai,Xiongkuo Min
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Aligning with the human visual system~(HVS) in perceiving and evaluating the quality of visual signals is a central objective of machine-vision-based visual quality assessment systems. With the rapid progress of large multi-modal models~(LMMs), visual question answering provides a promising paradigm for building unified foundation models for visual quality assessment under multi-modal and multi-task scenarios. Inspired by the classical perception-decision" process in HVS-based quality evaluation, we formulate visual quality assessment for LMM-based machine vision as two complementary tasks: quality interpretation’’ and ``quality scoring". Centered on these objectives, we propose LLaVA-Assessor, a unified data construction and model training system. To support multi-modal inputs, we design an adaptive model architecture that enables efficient processing of both images and videos. For data construction, we develop rigorous human annotation protocols and a novel machine-synthesis-dominated data expansion pipeline to build a large-scale and high-quality datasets. Furthermore, we introduce a simple yet effective prompt disentanglement strategy to alleviate training-objective confusion in multi-task learning, thereby enabling stable and coherent joint training. The resulting all-in-one LMM LLaVA-Assessor-GIGA achieves superior performance on 11 image/video quality scoring test sets and 4 visual quality interpretation benchmarks. Extensive results demonstrate the effectiveness of integrating structured data construction, adaptive model design, and multi-task joint training for automated visual quality assessment. Our work provides compelling insights for developing foundation LMMs for automatic visual quality assessment. Project page at this https URL.
[CV-43] opology-Aware Parameter-Efficient Adaptation for Cross-Dataset Retinal Vessel Segmentation
链接: https://arxiv.org/abs/2609.26189
作者: Yongsong Huang,Tomo Miyazaki,Kai Xu,Xiaofeng Liu,Yaohou Fan,Shinichiro Omachi
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: This manuscript is currently under peer review. Copyright may subsequently be transferred to the publisher, after which the availability of this version may be subject to the publisher’s policy
Abstract:Retinal vessel segmentation in multi-domain deployment requires a source model to adapt to domains that differ in imaging conditions and annotation conventions. Conventional parameter-efficient fine-tuning reduces target-specific storage, but its highly restricted adaptation subspace can be insufficient for reconstructing thin, connected vascular structures. We therefore ask how target-specific capacity should be allocated so that topology-aware supervision remains effective under a strict per-domain parameter budget. Based on this principle, we propose TAPDecoderFT, a topology-responsive, role-structured adaptation framework. Specifically, TAPDecoderFT shares a fixed source parameter state across deployment domains, uses low-rank residuals for target-specific private/fusion feature mixing, and retains a trainable dense-reconstruction path comprising the decoder, output head, and refinement module. To promote structurally faithful predictions, the compact target state is jointly optimized with a region-overlap and topology-aware objective that encourages centerline continuity and thin-branch recovery. It improves both DSC and clDice over GenericLoRA-r4 and narrow TAP-r4 in all six directions and is comparable to full fine-tuning.
[CV-44] End-to-End Visual Odometry with RNNs and Attention
链接: https://arxiv.org/abs/2609.26188
作者: Ruiyu Li,Yinjia Liu,Alexander Yu
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Video Odometry (VO) is the process of estimating the ego-motion of an object by analyzing visual information such as a sequence of frames from one or multiple cameras. It has been a popular research topic in computer vision and robotics, and its applications include mobile robotic systems as well as autonomous driving. In this project, we investigate existing end-to-end deep-learning approaches to VO, and propose a novel temporal attention-based model to improve upon the baseline. In addition, while the vast majority of existing deep-learning-based approaches to VO are trained on driving data, we investigate the performance of deep-learning-based VO to the more dynamic and complex problem of hand-held cameras.
[CV-45] MGRL-RSCC: Multi-Granularity Reward Reinforcement Learning for Fine-Grained Remote Sensing Change Captioning
链接: https://arxiv.org/abs/2609.26166
作者: Futian Wang,Mengqi Wang,Xiao Wang,Wentao Wu,Haowen Wang,Zhicheng Zhao,Jin Tang
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
备注:
Abstract:Remote Sensing Change Captioning (RSCC), which aims to generate accurate and detailed linguistic descriptions of ground object variations from bi-temporal remote sensing images, is a critical and challenging task in intelligent remote sensing interpretation. The mainstream autoregressive training paradigm faces severe exposure bias and train-test distribution mismatch, resulting in cumulative generation errors. They tend to produce conservative and template-fixed captions while ignoring subtle scene change details. To address these challenges, this paper proposes a novel multi-granularity reward reinforcement learning paradigm, termed MGRL-RSCC. Specifically, we first leverage a CNN and hierarchical self-attention module to extract and enhance visual features from bi-temporal remote sensing images. A Transformer decoder is then utilized to complete visual-to-linguistic translation. Different from existing methods, we design a dual-decoding strategy and a two-stage joint optimization scheme, which combines token-level supervised learning via greedy decoding and multi-granularity reward-driven self-critical reinforcement learning via sampling decoding. We further construct three complementary reward functions covering linguistic fluency, change state consistency, and structural-semantic relevance to comprehensively optimize caption quality and alleviate false and missing change descriptions. Extensive experiments on multiple public RSCC benchmark datasets demonstrate that the proposed MGRL-RSCC effectively mitigates exposure bias and conservative generation problems in traditional autoregressive methods. The source code and pre-trained models will be released on this https URL
[CV-46] Moving6DPoSe: A Multimodal Database for Monocular 6D Pose Estimation and Segmentation of Moving Objects
链接: https://arxiv.org/abs/2609.26161
作者: Ignacio Bugueno-Cordova,Javier Ruiz-del-Solar,Rodrigo Verschae
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Estimating the 6D pose of moving objects remains challenging due to motion blur and the limited temporal resolution of conventional frame-based cameras. Existing event-based datasets further provide limited sensing modalities, annotations, and motion scenarios. We introduce Moving6DPoSe, a multimodal database comprising two complementary subsets: Moving6DPoSe-R with real-world recordings and Moving6DPoSe-S with synthetic sequences generated from the same objects. The dataset contains 16 scanned objects and 1,702 real and synthetic rosbags spanning multiple motion scenarios, with annotations for semantic segmentation, object detection, and monocular 6D pose estimation. We further provide baseline results for all three tasks across frame and event-based modalities. Experimental results show that event-based representations achieve more robust moving-object segmentation than conventional RGB images, while monocular orientation estimation remains challenging, highlighting the potential of Moving6DPoSe for moving-object perception research.
[CV-47] Vorch-Human: Unified Multi-Task Human-Centric Generation via Long-Horizon Continuation
链接: https://arxiv.org/abs/2609.26117
作者: Yang Ding,Haoran Yu,Xin Ma,Yulei Lu,Menglin Han,Yaole Wang,Siqian Yang,Gang Yue,Kaihao Zhang,Yaohui Wang,Lin Ma
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: Project page: this https URL
Abstract:Human-centric audio-visual generation spans several closely related tasks: animating a person from driving speech, jointly generating speech and video from a voice reference, and synthesizing a scene from paired appearance and voice references. Existing systems commonly solve these tasks with separate models, even though they share the same target modalities and differ mainly in which observations are provided as conditions. We present Vorch-Human, a unified human-centric generation framework built on a dual-stream audio-video diffusion transformer. Vorch-Human augments the conventional noisy audio/noisy video interface with clean condition-audio and condition-video token groups. Per-token task embeddings, temporal position types, condition masks, and a shared multimodal prompt encoder allow driving speech, timbre examples, first frames, and subject images to be expressed within one model. To supply the supervision required by this interface, we develop a two-level data pipeline. Level 1 analyzes each clip with speech recognition, vocal separation, face detection and tracking, active-speaker and synchronization models, audio/visual speaker clustering, and multimodal caption correction; it produces subject-indexed speech, appearance, and timbre annotations. Level 2 links the same person across clips from a common source video and mines identity- and outfit-consistent reference images after face, body, quality, pose, and vision-language verification. Finally, we adapt Vorch-Human to long-form audio-driven generation by training with clean latent prefixes and using the same frozen-prefix recurrence at inference. Each segment contributes only its newly generated suffix, reducing boundary discontinuity and long-horizon identity drift. Experiments on short and five-minute generation demonstrate strong identity preservation, audio-visual synchronization, and temporal stability.
[CV-48] Neoadjuvant chemotherapy response prediction using pretreatment diffusion and contrast-enhanced magnetic resonance imaging with clinical variables
链接: https://arxiv.org/abs/2609.26105
作者: Pablo García Marcos,Paula Puerta González,Guillermo Lorenzo,Héctor Gómez,Covadonga del Camino,Adán Rodríguez,Ignacio Peláez,Angel Rio-Alvarez,Víctor M. González
类目: Artificial Intelligence (cs.AI); Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Prediction of pathological complete response before neoadjuvant chemotherapy may facilitate more tailored therapeutic planning for breast cancer patients. This work proposes a deep-learning model for pretreatment data only, combining apparent diffusion coefficient maps, dynamic contrast-enhanced magnetic resonance imaging, and clinical variables. The study uses the public ACRIN 6698/I-SPY2 multicenter dataset. The architecture employs EfficientNet-B0 pretrained encoders for image feature extraction and late fusion with clinical information. Multiple clinical variables were evaluated, including age, race, histological type, HR/HER2 subtype, SBR grade, and maximum diameter. Only HR/HER2 subtype improved the average area under the receiver operating characteristic curve (AUC) and was retained in the final model. Using stratified five-fold cross-validation, standalone apparent diffusion coefficient maps achieved a mean AUC of 0.79, whereas dynamic contrast-enhanced magnetic resonance imaging achieved 0.74. Adding HR/HER2 subtype improved performance to 0.83 and 0.81, respectively. The final configuration, using both imaging modalities and HR/HER2 subtype, achieved an AUC of 0.86. These results support pretreatment multimodal learning for response prediction, although external validation is required before clinical use.
[CV-49] MIAR: Medical Image Super-Resolution With Autoregressive Modeling
链接: https://arxiv.org/abs/2609.26103
作者: Fang Li,Yinglong Li,Hongyu Wu,Yang Gao,Minwei Zhao,Aimin Hao
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Medical Image Super-Resolution (MISR) aims to enhance spatial resolution without requiring hardware modifications. Although deep learning has yielded promising results, existing paradigms face a critical trade-off: diffusion-based methods suffer from prohibitive inference latency and compromised structural fidelity, whereas regression-based models typically produce over-smoothed results that lack perceptual realism. To address these limitations, we propose MIAR, which reformulates super-resolution as a conditional and progressive next-scale prediction task through a multi-scale autoregressive framework. To ensure structural fidelity, we augment the autoregressive backbone with a Scale-Adaptive Structural Decoder. Furthermore, we integrate a hierarchical beam search strategy during inference to mitigate the recursive error accumulation inherent in autoregressive generation, a phenomenon that is especially pronounced in medical images. Extensive experiments demonstrate that MIAR establishes new state-of-the-art benchmarks while maintaining superior fidelity. Notably, our framework achieves a 7.86% improvement in the perceptual metric MUSIQ compared with the state of the art, while simultaneously delivering a 2.02x speedup over diffusion-based methods.
[CV-50] st-time Reinforcement Learning for Anomalous Video Understanding
链接: https://arxiv.org/abs/2609.26099
作者: Huining Li,Yuxiang Duan,Jiyang Tan,Qian Li,MingCai Chen,Jian Zhang,Xingdong Sheng,Yuntao Du
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Anomalous video understanding aims to identify abnormal events in videos and interpret their semantic meanings beyond simple anomaly detection. Recent video large language models (Video-LLMs) have demonstrated promising zero-shot capabilities for this task, yet their performance remains limited due to insufficient adaptation to diverse anomaly patterns and evolving environments. Test-time reinforcement learning offers a promising solution by enabling models to improve through self-generated feedback signals without requiring additional human annotations. However, applying it to anomalous video understanding remains challenging due to three issues: (1) generated pseudo-labels can be unreliable when consensus is weak; (2) binary reward designs fail to capture uncertainty in model generations, resulting in ineffective optimization signals; and (3) unanimous rollout groups receive identical rewards, causing group-relative advantages to collapse and eliminating effective policy-gradient signals. To address these challenges, we present a novel test-time reinforcement learning framework for anomalous video understanding by introducing dual-query consistency filtering, an entropy-aware consensus reward, and a virtual negative anchor mechanism. The framework retains reliable samples through consistency across semantically equivalent queries, combines answer agreement with generation uncertainty for reward estimation, and introduces a virtual negative anchor to create reward variation in unanimous rollout groups, thereby preserving effective group-relative optimization signals. Experiments on VAU-Bench show that our method outperforms the compared frozen and supervised baselines. The gains are most pronounced on the ECVA subset of VAU-Bench with thinking, where accuracy improves from 75.81% to 90.00% relative to the frozen backbone.
[CV-51] FusionMMT: A Unified Multimodal and Multitask Learning Framework for Nuclear Fusion
链接: https://arxiv.org/abs/2609.26095
作者: Qiang Chen,Xiao Wang,Qingquan Yang,Hao Si,Zikang Yan,Meiwen Chen,Guosheng Xu,Jin Tang
类目: Artificial Intelligence (cs.AI); Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:With the growing global demand for energy, nuclear fusion has emerged as a promising direction for future clean energy. Tokamaks represent one of the leading approaches to magnetic-confinement fusion. Achieving high-performance, long-pulse, and steady-state operation requires effective diagnosis of plasma states. However, existing intelligent diagnostic methods are largely limited to either multimodal single-task or unimodal multitask learning, while a unified multimodal multitask learning framework remains underexplored. To address this gap, we construct EAST-VTD640, a multimodal multitask dataset that integrates vision and time-series diagnostics from 640 EAST shots for disruption prediction, edge-localized mode (ELM) recognition, and H98 regression. On this basis, we present FusionMMT, the first unified multimodal multitask framework for intelligent tokamak plasma diagnostics. FusionMMT employs multi-scale, time-aware, and variable-aware modeling to handle heterogeneous sampling rates and the high computational cost of high-frequency sequences. It further combines task-adaptive multimodal fusion with progressive multitask optimization to learn shared and task-specific representations while mitigating cross-task conflicts and optimization imbalance. Extensive experiments on EAST-VTD640 show that FusionMMT outperforms representative multimodal multitask methods across disruption prediction, ELM recognition, and H98 regression. The source code will be released on this https URL
[CV-52] RECAP: Relation Evidence Calibration for Detecting Spatial Relation Hallucinations in Vision-Language Models
链接: https://arxiv.org/abs/2609.26093
作者: Feixiang Liu,Qiang Qiu,Qingyang Li,Hui Xu
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: 21 pages, 4 figures, including supplementary material. Code: this https URL
Abstract:Vision-language models can answer spatial relation questions confidently even when the image supports an incompatible relation. We formulate relation-grounded selective prediction: accept or reject an already-produced yes/no answer by auditing its visual support, rather than treating uncertainty as evidence. RECAP, our relation-evidence calibration framework, compares image-conditioned likelihoods for a claim, its semantic contradictions, and optional one-sided supports, then converts these witnesses into an answer-conditioned rejection risk. A calibration-only gate preserves confidence as a veto when confidence is demonstrably informative and otherwise deploys relation evidence alone. Across 20 group/image-disjoint splits, RECAP lowers H-FPR@80 over confidence by between 2.0 and 17.9 points on VSR and raises Acc@80 by 3.0, 8.6, and 12.6 points on What’sUp for Qwen3-VL-8B, InternVL3.5-8B, and LLaVA-1.5-7B. It outperforms matched VCD-style visual contrast on all four primary metrics in all six settings. Full-pool VSR fallback, target-ranked GSR-Bench transfer, equal-budget supervised controls, and two additional checkpoints show a consistent operating principle: structured counterevidence complements certainty when confidence is misaligned, while the gate retains confidence when it is already useful.
[CV-53] Match One Learn with Graph: One-to-Graph Query Collaboration with Backward Sharing for Object Detection
链接: https://arxiv.org/abs/2609.26092
作者: Wenxiao Fan,Jingling Fu,Luohang Liu,Lichen Ma,Yu He,Zhiyang Yu,Weishan Bi,Junshi Huang,Yan Li,Gu Simiu,Kan Li
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: preprint
Abstract:One-to-one (O2O) matching enables Detection Transformers (DETRs) to perform end-to-end set prediction by assigning each object to a single positive query. However, the strongest classification, center, scale, and overlap evidence for an object is often distributed across multiple queries. This mismatch leaves only the matched owner positively supervised for the object, while other evidence-bearing queries receive no box target for it. We term this query knowledge fragmentation. To exploit such complementary evidence without one-to-many supervision, we propose BS-O2G, a plug-in that builds a sparse prediction-aware graph from decoded features, boxes, and class distributions to organize query collaboration in feature and optimization spaces while preserving the original O2O matcher, positive labels, and objective. One-to-Graph (O2G) calibration propagates relative messages over this graph to consolidate query evidence in the forward pass, whereas Backward Sharing (BS) reuses its transposed detached adjacency to route gradients across persistent query basis vectors without changing the decoder input in the forward pass. Experiments across diverse DETR methods, backbones, COCO, and CrowdHuman show consistent gains and faster convergence with negligible parameter/FLOP growth and modest runtime overhead, supporting graph-based query collaboration as an alternative to expanding positive assignments.
[CV-54] BDSLI: A hybrid CNN-Transformer model for Bengali Sign Language interpretation
链接: https://arxiv.org/abs/2609.26088
作者: Abir Bin Yousuf,Muhammad Iqbal Hossain
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:This study introduces a novel hybrid CNN-Transformer architecture to address the limited progress in Bengali SLR, focusing on isolated sign word recognition and sentence generation. This specific model combination is new to Bengali SLR tasks. A custom video dataset was developed, featuring 62 distinct Bengali sign words (250 samples/class), along with a separate test dataset. The CNN-Transformer model demonstrated superior performance against all comparative and baseline models (e.g., CNN-LSTM, standalone TCN), achieving a 99.58% training accuracy (99.48% validation) and a 98.65% test accuracy. The trained model was subsequently deployed in a web application for real-world validation.
[CV-55] Margin-Drop Coordinates for Cross-Budget Robustness Evaluation
链接: https://arxiv.org/abs/2609.26081
作者: Yanliang Huang,Zhen Zhang,Peng Xie,Wenyuan Wu,Sitong Zhu,Zhuoqi Zeng,Amr Alanwar
类目: Machine Learning (cs.LG); Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Fixed-budget robustness evaluation can select the wrong frozen vision encoder. An encoder that survives a shallow attack may lose most of that robustness when the same evaluation is strengthened. We ask whether the shallow evaluation contains enough information to identify this budget fragility. For each clean-correct sample, the evaluation records the clean pairwise margin, the first-order linearized margin-drop scale, the margin drop from a clean-start one-step attack, and the drop reached by an iterative attack. Normalizing by that scale gives three margin-drop coordinates capturing clean margin slack, one-step shortfall, and drift, where drift is the additional normalized margin drop the iterative attack reaches beyond the one-step perturbation. Together, they reconstruct the normalized post-attack margin and therefore the pass-or-fail outcome. Across 42 pretrained frozen vision encoders, the shallow survival rate carries essentially no rank information about subsequent PGD-10 to PGD-200 collapse, at Spearman -0.006, while the median shallow drift coordinate ranks the same collapse at +0.811. The result persists in a held-out encoder pool and under an \ell_\infty evaluation. With deep evaluation limited to 11 encoders, ranking by shallow drift recovers 11 of the 17 high-collapse encoders, compared with 5 under survival-rate ranking. The full coordinate decomposition further distinguishes cases that share the same fixed-budget residual but diverge at deeper budgets, and separates margin repair from drift repair under interventions, revealing distinct repair paths that endpoint robustness alone does not identify.
[CV-56] oW3D: Consistency-aware Interactive Point-based Mesh Editing on GANs
链接: https://arxiv.org/abs/2609.26078
作者: Haixu Song,Fangfu Liu,Chenyu Zhang,Yueqi Duan
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:In this paper, we propose ToW3D that enables precise and consistent control over 3D generative adversarial networks (GANs) with the Tug-of-War competition between shape deformation and appearance consistency. Existing point-based GAN editing methods such as DragGAN and GANWarping have yielded impressive performance for 2D image manipulation. However, as 3D generators present weaker generalization ability compared with 2D due to limited training data, they would suffer from drastic changes in global appearance when editing local areas of meshes. To address this, we design a pipeline of ``drag locally, shove globally’', which iteratively performs two optimization steps: 1) pull the point towards the target, and 2) push the structure and semantics back to the source. Specifically, we design a structure adaption module based on structure which guarantees the preservation of basic geometric properties, and a semantic preservation module that maintains semantic similarity across different views. Extensive qualitative and quantitative experiments demonstrate superiority of our ToW3D approach over prior methods in terms of appearance consistency and fidelity especially under large deformations.
[CV-57] Cellular-Communication-Level Interpretability for Pathology Foundation Models via Graph Distillation on Microenvironment
链接: https://arxiv.org/abs/2609.26073
作者: Yuxiang Xiao,Zhiwei Chen,Dan Dai,Wei Li,Tianyang Zhang,Yakun Ju,Yang Hu,Kaixiang Yang
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: 11 pages, 3 figures, 1 table
Abstract:Pathology foundation models (PFMs) provide strong tile-level representations but remain difficult to interpret at the cellular and microenvironmental scales that underpin clinical reasoning. We introduce Graph-Interpreter (G-Interp), a graph-distillation framework that equips a frozen PFM teacher with a cellular-communication-level “plug-in” interpreter, without modifying the teacher. For each tile, we segment cells as graph nodes and construct a microenvironment graph based on spatial adjacency. Graph neural network (GNN) students distil the PFM embedding, whilst learning attention-based message passing that yields node- and edge-level importances. We interpret these importances as cell-cell communication evidence, providing fine-grained explanations of how PFMs encode microenvironmental context. To stabilise distillation when graph abstraction is imperfect, we employ a lightweight auxiliary student to supply complementary visual cues and condition graph message passing, while keeping the primary interpretability signal graph-derived. We evaluate explanation faithfulness by mapping graph-selected evidence back to the image using instance masks and measuring teacher sensitivity under targeted vs non-target occlusions. Across multiple histopathology tasks, G-Interp produces highly scalable, microenvironment-aware explanations, while maintaining competitive predictive performance.
[CV-58] SPEANet: Structural Prior Enhanced Attention Network for Parameter-Efficient Remote Sensing Object Detection
链接: https://arxiv.org/abs/2609.26064
作者: Wei Lu,Junjie Li,Feifei Sang,Si-Bao Chen
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Remote sensing object detection (RSOD) requires compact backbones capable of preserving weak geometric cues under extreme scale variation and background clutter. Fixed structural operators provide complementary contour and frequency responses without introducing learnable operator coefficients. However, directly injecting these responses can amplify content-irrelevant textures, while applying a uniform operator design across the hierarchy may be poorly matched to stage-specific representation requirements. We propose the Structural Prior Enhanced Attention Network (SPEANet), a parameter-efficient RSOD backbone that integrates fixed operators through stage-specific prior extraction and context-conditioned response modulation. SPEANet assigns smoothed contour and multi-order directional modeling to shallow, high-resolution features, while employing a compact approximation-detail interaction mechanism in deeper stages. Learned spatial gates regulate the resulting prior responses before residual fusion. Experiments on five benchmarks, together with evaluations across seven detection frameworks on DOTA-v1.0, achieve a favorable accuracy-parameter trade-off. With Oriented R-CNN, SPEANet achieves 78.55% mAP on DOTA-v1.0, 72.24% mAP on DOTA-v1.5, and 67.30% mAP on DIOR-R using 23.0M total parameters, including a 5.97M-parameter backbone.
[CV-59] CricRAG : Retrieval Augmented Vision-Language Models for Personalized Cricket Coaching AAAI25
链接: https://arxiv.org/abs/2609.26056
作者: Agamdeep Singh,Sujit PB,Mayank Vatsa
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
备注: AAAI 25 - Towards Knowledgeable Foundational Models workshop
Abstract:Vision-Language Models (VLMs) offer promising capabilities for automated sports coaching but face a fundamental limitation: they implicitly compare against professional standards, making their feedback impractical for developing players. We present CricRAG, a retrieval-augmented framework that aligns VLMs with skill-appropriate benchmarks for personalized cricket coaching. Our key insight is that by retrieving similar-but-better techniques as reference points, we can guide VLMs to provide developmentally appropriate feedback that mirrors human coaching practices. We contribute: (1) a labelled dataset of 288 cricket technique videos spanning multiple skill levels, (2) an efficient motion retrieval pipeline using contrastive learning that achieves 78% top-3 retrieval accuracy, (3) a frame sampling technique that reduces inference costs, and (4) a retrieval-augmented approach that significantly improves feedback alignment with coaching principles, achieving up to 94% agreement with professional assessments compared to 67% without retrieval context.
[CV-60] EMERGE: Resolution-Agnostic Point Cloud Generation with Equivariant Graph-Based Diffusion
链接: https://arxiv.org/abs/2609.26039
作者: Ilias Mitsouras,Nikolaos Chaidos,Giorgos Stamou,Athanasios Voulodimos
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
备注: 26 pages, 11 figures
Abstract:Point cloud generation has emerged as a crucial task for accurately capturing and reproducing the complexity of the physical world. However, existing generative approaches, predominantly relying on Transformers and Variational Autoencoders (VAEs), frequently ignore the continuous, non-grid topologies inherent to 3D spaces. Although the integration of graph-based structures has yielded significant benefits in related discriminative vision tasks, such geometric architectures remain noticeably absent from 3D generative modeling. To address this gap, we introduce EMERGE (Equivariant Multi-scale GNN for Resolution-agnostic point cloud GEneration), the first fully SE(3) -equivariant graph-based diffusion backbone explicitly designed to generate point clouds while preserving continuous spatial symmetries. Our framework bypasses the rigid resolution dependencies of standard generative pipelines, enabling zero-shot inference at multiple, arbitrary spatial resolutions. Extensive empirical evaluations demonstrate that EMERGE achieves State-of-the-Art generation quality across standard metrics, while the strong inherent geometric inductive biases enable significantly faster training convergence compared to existing baseline methods.
[CV-61] NAWE: Digital Watermarking with Neural-Assisted Watermark Extraction
链接: https://arxiv.org/abs/2609.25972
作者: Roman Chaban,Vitaliy Kinakh,Lilian Rouzaire,Slava Voloshynovskiy
类目: Computer Vision and Pattern Recognition (cs.CV); Multimedia (cs.MM)
备注:
Abstract:NAWE (Neural-Assisted Watermark Extraction) combines an explicit signal-processing watermarking construction with a pretrained neural host predictor. A periodic, perceptually masked watermark carrier provides synchronization, Polar coding supplies redundancy, and denoising followed by subtraction extracts the embedded watermark. The denoiser remains frozen, without watermark-specific training. A one-factor-at-a-time study compares Wiener, BM3D, DRUNet, and GS-DRUNet host estimators. Comparisons with TrustMark, SSL Watermarking, PixelSeal, and WAM show NAWE’s lowest geometric and photometric class BER and strong message recovery, while filtering and noise remain limitations consistent with the non-adaptive selection of the watermark extractor. The comparison retains the systems’ different payloads and coding.
[CV-62] GRIP: Gaussian Rendering as a Cross-Modal Bridge for Image-to-Point Cloud Registration
链接: https://arxiv.org/abs/2609.25966
作者: Karim Slimani,Catherine Achard,Eric Marchand,Brahim Tamadazte
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:This paper introduces GRIP, a pose-conditioned refinement framework for pixel-to-point matching and 2D to 3D registration. Given an initial coarse pose estimate, GRIP addresses the structural mismatch between grid based image descriptors and unordered point cloud descriptors by softly rendering learned 3D point features onto the image grid through Gaussian feature splatting. The rendered point derived feature map is then fused with image features by a pixel aligned transformer, enabling visual semantic and geometric cues to interact in a shared 2D representation. The refined features are decoded and propagated to finer resolutions for dense correspondence estimation and final pose refinement. Experiments on RGB D Scenes V2 and 7 Scenes demonstrate state of the art inlier ratio and competitive registration recall, with stronger performance under stricter evaluation thresholds.
[CV-63] owards Systematic Qualification of Vision-Language Models for Automotive Perception Systems
链接: https://arxiv.org/abs/2609.25945
作者: Malsha Ashani Mahawatta Dona,Konstantinos Rokanas,Alexander Säfström,Krishna Ronanki,Christian Berger
类目: Computer Vision and Pattern Recognition (cs.CV); Software Engineering (cs.SE)
备注: Accepted in ICTSS 2026 - 38th International Conference on Testing Software and Systems
Abstract:The field of Artificial Intelligence has been adopted for many application domains. Vision Language Models are one of the recently advanced AI techniques that have been explored to support automotive features such as vehicle perception, and safety assurance. However, such language models are prone to hallucinations, posing a potential threat to the safety of automotive systems that may incorporate them. Within the automotive domain, VLMs could not only hallucinate traffic objects, but could also fail to identify traffic objects that are actually present, which may potentially lead to dangerous situations. Though we have observed a growing body of literature that proposes verification and validation techniques for safe and trustworthy AI, these methods are often studied in isolation, focusing either on run-time or design-time phases. Such isolated techniques could be insufficient in safety-critical, realistic contexts such as automotive perception systems. In this paper, we analyze design-time and run-time verification and validation techniques based on a taxonomy presented by Huang et al. We present an automotive study in which a design-time qualification workflow is proposed to complement run-time monitoring. This workflow combines a fixed safety-relevant ontology-based structured annotation system together with a synonym-based evaluation process to statistically evaluate three state-of-the-art VLMs against data from the nuScenes dataset. We observed that the proposed technique enables deterministic and repeatable quantification of the hallucinations VLMs generate in automotive perception-related tasks. The proposed workflow supports model comparison and deployment-oriented engineering decisions within the design-time verification and validation process and will contribute to a holistic verification strategy that strives towards trustworthy automotive perception systems
[CV-64] Calibrating Retrieval Geometry: Reliability-Guided Training-Free Aggregation for Visual Place Recognition
链接: https://arxiv.org/abs/2609.25937
作者: Xin Li,Zhimin Mao,Shang Wang,Siyuan Duan,Geng Zhang
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: 26 pages, 5 figures, 9 tables, including appendices
Abstract:Frozen visual foundation models provide transferable features for visual place recognition, but fixed aggregation can suppress useful distinctions in new environments. We introduce TFA, a reliability-guided, training-free aggregation method requiring neither place labels nor task-specific weight updates. Our key observation is that reproducible retrieval need not be discriminative: independent codebooks can consistently retrieve a few database hubs. TFA combines cross-codebook agreement, retrieval coverage, and spectral statistics to control residual assignment, spectral shaping, and global-feature fusion. Its spectral kernel exactly recovers original descriptor similarity at zero intervention. Database-only TFA fixes its rules before accessing queries; TFA-C64 uses 64 disjoint unlabeled target images to calibrate retrieval for subsequent queries. Across 20 ground protocols with a fixed DINOv2-B backbone and matched resolution, database-only TFA improves Recall@1 over AnyLoc by 17.39 percentage points on MSLS-val and 9.55 on SPED. C64 mitigates failures of database-only calibration in driving environments. Across eight aerial/cross-view protocols, TFA achieves the highest Recall@1 among compared training-free heads in 14 of 16 DINOv2/DINOv3 backbone-protocol combinations. In a separate native-system comparison, DINOv2-G-based TFA-C64 reaches 91.46% Recall@1 on Pitts30k and 76.29% on VPAIR, outperforming the displayed training-free comparators on all five benchmarks. These results show that reliability-guided aggregation can recover additional retrieval capability from frozen representations, providing a practical baseline for new environments with scarce place supervision.
[CV-65] AT3D-AD: Anomaly Type-Aware 3D Anomaly Detection via Hierarchical Point-Language Alignment
链接: https://arxiv.org/abs/2609.25930
作者: Jingyu Zeng,Haoquan Lu,Can Gao
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Detecting and localizing 3D point-cloud defects is essential for industrial inspection. However, existing methods often suffer from imprecise localization due to the lack of anomaly supervision and reliance on single-granularity representations. To address these limitations, we propose Anomaly Type-Aware 3D Anomaly Detection (AT3D-AD), a unified framework for joint detection, localization, and classification. Specifically, we first design the Physics-Driven Parametric Anomaly Synthesis (PDPAS) module employing multiple parametric functions to generate synthetic anomalies, providing explicit anomaly supervision. Then, we propose the Hierarchical Global-Local Anomaly Alignment (HiGLA) module to align global and local representations within the normal and anomalous groups. Finally, we propose the Semantic-Geometric Anomaly Classification (SGAC) module to jointly learn localization and classification, yielding spatially precise and type-discriminative anomaly representations. Extensive experiments establish new state-of-the-art performance on all four benchmarks. AT3D-AD achieves Object/Point AUROC scores of 98.1%/98.9% on Anomaly-ShapeNet and 95.0%/95.2% on Real3D-AD, while reaching 74.2% Macro-F1 for anomaly-type recognition on Real3D-AD.
[CV-66] NaCR: Visual Localization via NeRF-aided Camera Ray Regression
链接: https://arxiv.org/abs/2609.25907
作者: Yesheng Zhang,Xiang Dai,Xu Zhao,Chongyang Zhang
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: v0
Abstract:Visual localization (VL) is a fundamental technology for vision applications such as virtual reality. Recently, a novel VL paradigm, Camera Ray Regression (CRR), has emerged, which maps 2D image patches to 3D camera rays, but its accuracy is limited. To improve CRR accuracy, we notice a compelling duality: the inverse of this mapping is inherently performed by the novel view synthesis model, \ie, Neural Radiance Fields (NeRF). While NeRF renders image patches from camera rays via differentiable ray marching, CRR predicts the rays from image patches. Motivated by this complementary relationship, we propose NeRF-aided Camera Ray Regression (NaCR), a unified framework that seamlessly bridges NeRF and CRR at the ray level. First, NaCR incorporates three simple yet effective enhancements into the CRR baseline. Second, leveraging a pre-trained NeRF, NaCR augments the training data by synthesizing novel views tailored for efficient, patch-level consumption. Finally, exploiting the differentiability of NeRF, NaCR forms a closed-loop supervision pipeline where photometric rendering errors are back-propagated to optimize the predicted camera rays. To ensure stable convergence within the highly non-convex image space, we introduce a two-stage training curriculum. Extensive experiments across indoor and outdoor benchmarks demonstrate that NaCR achieves competitive accuracy. Comprehensive ablation studies validate the efficacy of each proposed component.
[CV-67] BAS-OPD: Budget-Aware Selective On-Policy Self-Distillation for Fine-Grained Multimodal Perception
链接: https://arxiv.org/abs/2609.25891
作者: Zihan Chen,Hengguang Zhou,Yuan Kang,Yiming Zhang,Wenhui Fang,Zenghui Ding,Yining Sun,Cho-Jui Hsieh
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
备注:
Abstract:Multimodal large language models (MLLMs) often struggle with fine-grained visual perception when processing complete images, as critical evidence may only appear in local regions. On-policy self-distillation (OPD) enables transferring privileged visual knowledge from informative views to full-image policies, but querying the teacher for every rollout introduces substantial supervision costs. In this work, we propose BAS-OPD, a budget-aware selective OPD framework that allocates teacher supervision under limited query budgets. Instead of querying all rollouts, BAS-OPD selects informative samples while maintaining full-batch student generation. We explore random, uncertainty-based, and learned utility-based selection strategies, where the learned selector estimates query value from detached rollout statistics and online utility signals derived from student–teacher agreement and teacher confidence without additional student forward passes. BAS-OPD only changes training-time supervision allocation and preserves single-pass full-image inference. Experiments on fine-grained multimodal perception benchmarks demonstrate that BAS-OPD achieves strong performance while substantially reducing teacher supervision costs, highlighting the effectiveness of selective OPD under constrained budgets.
[CV-68] LoRango: It Takes Two LoRAs to Unlock Hidden Behaviors in Diffusion Models
链接: https://arxiv.org/abs/2609.25884
作者: Jin Wei,Rundong Li,Ruihao Yang,Yikai Wang,Xiaoyuan Duan,Jianxiong Wu,Yanbo Wang,Chang Xu,Lingyun Zhang,Zhuyang Yu,Ping Chen,Jun Dai,Xiaoyan Sun
类目: Cryptography and Security (cs.CR); Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Users commonly combine multiple Low-Rank Adaptation (LoRA) adapters to personalize images with different subjects, styles, and visual attributes. Yet inspecting adapters individually does not establish the safety of their composition. We identify and characterize a pair-conditioned attack in text-to-image diffusion: individually useful and benign-appearing adapters redirect image generation when co-loaded with a specifically matched partner, whose identity serves as the trigger. We introduce LoRango to realize this attack through complementary Signature and Payload adapters. The Signature writes a pair-specific code into intermediate carrier representations, while the Payload uses code-selective responses and opposing signal/reference branches. These branches approximately cancel for standalone adapters and mismatched pairs; matched code-reader alignment breaks cancellation within native GEGLU blocks and releases the programmed action. Both adapters are exported as ordinary static LoRA files compatible with standard loaders, requiring no prompt trigger or base-pipeline modification. LoRango achieves matched-pair attack success rates of 97.9% on SD v1.5 and 98.7% on SDXL, compared with 2.8–4.6% when implanted adapters are loaded individually. Further experiments evaluate pair selectivity, standalone fidelity, robustness to deployment variations, and applicability across denoiser architectures. These findings show that individual-adapter inspection is insufficient to assess the security of multi-LoRA personalization and motivate auditing adapter compositions.
[CV-69] Delving into Asymmetric Information Dynamics for High-Fidelity Virtual Try-On
链接: https://arxiv.org/abs/2609.25881
作者: Zishu Qin,Zhiyu Jin,Pipei Huang,Hao Zhou
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Virtual try-on (VTON) requires precise pixel-level fidelity, yet mainstream Diffusion Transformers (DiTs) often suffer from texture degradation and structural drift. We identify symmetric interactions in standard joint-attention mechanisms as a source of these failures. Although such interactions support semantic flexibility in general-purpose editing, they allow stochastic noise to corrupt deterministic garment features in VTON. We analyze this problem through asymmetric information dynamics and introduce two diagnostic indicators: Conditional Attention Entropy (CAE) for feature unbiasedness and Injected Information Flux (IIF) for injection effectiveness. Our analysis suggests that symmetric bidirectional attention can corrupt conditional features and attenuate the conditional signal. To address these limitations, we propose RealFit, a framework that combines Unidirectional Information Flow (UIF) with Decoupled Timestep Modulation (DTM). UIF isolates the garment condition from stochastic noise to preserve garment identity, while DTM optimizes the modulation scale to maintain a strong conditional signal. The resulting time-invariant condition branch enables a conditional KV cache that reduces inference time by approximately 75%. RealFit offers a principled approach to conditional generation with state-of-the-art fidelity and efficiency.
[CV-70] V-AudioRemover: Joint Text-Visual Guided Sound Removal with Multi-Task Hard-Mixture Curriculum
链接: https://arxiv.org/abs/2609.25864
作者: Xinyue Guo,Jianxuan Yang,Daiguo Zhou,Jiagao Hu,Yuxuan Chen,Fei Wang,Jian Luan
类目: Multimedia (cs.MM); Computer Vision and Pattern Recognition (cs.CV); Sound (cs.SD)
备注:
Abstract:Visual object removal can eliminate a target from video frames, yet its acoustic trace persists in the soundtrack, causing obvious audio-visual inconsistency. Existing video inpainting models operate solely on pixels, while audio editing models, especially for the sound removal task, are typically driven by text and therefore rely on limited single-modal control, which is less effective than multimodal guidance that provides stronger semantic grounding and temporal synchronization cues. In this paper, we present Text-Visual Guided Sound Removal (TV-AudioRemover), a target sound removal framework that leverages the visually edited video together with a natural-language instruction to suppress the sound associated with the removed visual object from the original audio mixture. To acquire high-quality training data, we devise a pipeline to construct a million-scale dataset of single-object audio-visual aligned samples, from which we synthesize mixture-target pairs customized for model training. To effectively leverage visual context and follow instruction intent, we augment the model architecture with task tokens, generalizable instruction modeling, and modality-specific global guidance. We further adopt multi-task training to strengthen task-role comprehension, and employ a hard-mixture curriculum that leverages semantically similar acoustic mixtures during fine-tuning to enhance fine-grained source discrimination. To support evaluation, we present AV-Remove-Bench, a comprehensive audio-visual object removal benchmark, along with dedicated objective metrics and an MLLM-based evaluation protocol. Experiments demonstrate that our method achieves state-of-the-art performance on both subjective and objective metrics. Project page: this https URL.
[CV-71] MatchFusion: Explicit-Implicit Instance Matching for Spatio-Temporal Multimodal Autonomous Driving
链接: https://arxiv.org/abs/2609.25860
作者: Xiaoyu Li,Jiajia Fu,Long Shi,Tianyu Du,Ruihang Li,Xian Wu,Lijun Zhao,Yingtao Zhang,Lining Sun,Ruifeng Li
类目: Computer Vision and Pattern Recognition (cs.CV); Robotics (cs.RO)
备注: 8 pages, 4 figures
Abstract:Sparse instance representations provide a compact interface for spatial LiDAR-camera and temporal past-current interaction in multimodal perception and E2EAD. Effective interaction requires reliable instance correspondences despite geometric discrepancies and heterogeneous semantic representations. Attention-based methods exploit contextual semantics but often require specialized representation alignment, increasing computational overhead. In contrast, association based on structured object states is efficient and interpretable but lacks contextual evidence to resolve ambiguous matches. To combine these complementary strengths, we propose MatchFusion, a learnable instance matching and fusion module for spatio-temporal multimodal autonomous driving. MatchFusion initializes pairwise affinities using geometric similarity and category consistency, then selectively refines structurally plausible associations using instance embeddings. The resulting soft matchmap guides a common residual aggregation operator for adaptive information exchange. This unified matching-fusion formulation supports spatial LiDAR-camera and temporal past-current interaction, using multi-view image-plane geometry and motion-compensated BEV geometry as the respective structural priors. Experiments on nuScenes demonstrate consistent perception gains across diverse front-end configurations. Compared with a prior instance-centric fusion method, the MatchFusion-equipped system achieves higher perception accuracy while reducing FLOPs by 55.3% and GPU memory usage by 39.3%, with the matching-fusion module accounting for only 3.7% of total perception latency. Integrating temporal MatchFusion into SparseDrive further improves perception within an E2E framework without additional supervision. These results establish explicit-implicit matching as an effective and efficient mechanism for spatio-temporal instance interaction.
[CV-72] Less Is More in the Long Tail: Stage-Adaptive Sample Selection for Annotation-Efficient Dense Prediction
链接: https://arxiv.org/abs/2609.25850
作者: Xiaofei Du,Lei Zhang,Shuyu Yan,Manning Wang,Zhijian Song
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Deep learning performance generally improves with increasing training data, yet this scaling is fundamentally constrained by annotation cost in large-scale dense prediction tasks with long-tailed category distributions, where pixel- or voxel-level annotation is prohibitively expensive. We propose SASS (Stage-Adaptive Sample Selection), a stage-adaptive data-selection framework for pool-based active learning in long-tailed dense prediction. SASS combines three components: label-free self-supervised gradient scoring, prior-guided category rebalancing with validation-driven feedback, and stage-adaptive acquisition aligned with model training dynamics. This design avoids candidate ground-truth masks during gradient scoring while making acquisition responsive to long-tail imbalance and evolving representations. We evaluate SASS on a multimodal 3D medical segmentation testbed comprising over 100,000 samples spanning 108 anatomical structures. SASS recovers 98.3% of full-dataset performance with a 40% training-pool annotation budget, outperforming BADGE by 5.1 percentage points. Moreover, SASS exhibits a statistically supported less-is-more pattern, surpassing full-dataset training at the Hard-group level and, at the structure level, for the pancreas and gallbladder. More broadly, SASS shows that annotation-efficient learning depends not only on which samples are selected, but also on how the annotation budget is distributed across categories and when model-derived scores begin to guide selection.
[CV-73] Visual Jev: Accurate and Efficient Decisions from Shared Visual Context
链接: https://arxiv.org/abs/2609.25845
作者: Guanxu Yu,Yuhang Yao
类目: Computer Vision and Pattern Recognition (cs.CV); Machine Learning (cs.LG)
备注: Code: this https URL
Abstract:Many vision applications ask several independent, forced-choice questions about the same image. Visual Jev encodes the image and public context once, executes isolated question suffixes as a batch, and reads candidate probabilities from the backbone’s language-model head. Across four benchmarks, answer-supervised post-training raises equal-weight macro accuracy from 70.6% to 76.1%, with the gain concentrated on the two task families represented in training. At N=32 questions per image, shared batched execution is 8.9x faster in warm amortized time than independent serial execution and remains 3.4x faster than an already-batched baseline that recomputes the prefix, at the cost of higher peak memory. A matched typed-head control offers no consistent accuracy advantage over the language-model-head readout. The supported design is therefore simple: adapt the backbone for quality, retain the existing readout, and share execution for efficiency.
[CV-74] Metric-Bench: Exploring In-context Spatial Metric Reasoning in VLMs for Indoor Scenes ECCV
链接: https://arxiv.org/abs/2609.25841
作者: Yuling Xi,Haokai Zhang,Muzhi Zhu,Hao Zhong,Zongze Du,Hengyu Zhao,Chenchen Jing,Yufei Yin,Bin Qin,Yongjie Yang,Zhenbo Luo,Hao Chen,Chunhua Shen
类目: Computer Vision and Pattern Recognition (cs.CV); Multimedia (cs.MM)
备注: Accepted to ECCV
Abstract:Metric reasoning is a critical and challenging task for Vision Language Models (VLMs), playing a pivotal role in embodied AI tasks such as robotic manipulation and autonomous navigation. However, current spatial reasoning remains bottlenecked by rigid pixel-level supervision; such localized optimization often compromises general multimodal intelligence, triggering performance degradation or catastrophic forgetting of broad reasoning capabilities. To address these limitations, we introduce Metric-Bench, a focused benchmark designed to guide metric-spatial reasoning using contextual information. By incorporating in-image reference objects with known physical dimensions, Metric-Bench guides models to implicitly learn the 2D-to-3D mapping without camera intrinsics. We further present MetricReasoner, a task-adapted reinforcement fine-tuning recipe for reference-grounded metric reasoning, using structured prompts and verifiable numerical rewards. Extensive experiments on Metric-Bench demonstrate that our approach significantly enhances spatial metric understanding, outperforming existing and even larger proprietary models by 43.1%, while improving downstream embodied performance over a spatial-specialized counterpart by 30.4% on RoboSpatial overall accuracy and 9.3% on ERQA, and additionally delivering consistent gains on general benchmarks (15.9% on V \star Bench, 88.9% on BLINK), indicating that the proposed adaptation does not necessarily compromise general VLM capabilities.
[CV-75] Identity-Centric Video Summarization via Hierarchical Fusion of Biometric Appearance and 3D Body Features
链接: https://arxiv.org/abs/2609.25837
作者: Milad Mirjalili,Enrique Alegre Gutiérrez,Eduardo Fidalgo Fernández,Víctor González Castro,Rocío Alaiz Rodríguez,Manuel Castejón Limas
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: 21 pages, 3 figures
Abstract:This work presents a video summarization algorithm based on multi-object tracking and person reidentification. We integrate facial embeddings, 3D body-shape features, and visual appearance into a unified tracking framework. These representations enable hierarchical identity assignment and tracking through bidirectional anchoring, which robustly recovers trajectories under severe occlusion or low visual quality. From these stable trajectories, we generate a compact set of summaries for each identity. We select keyframes using a multi-factor weighting scheme that optimizes biometric clarity, social interaction, and motion dynamics, while Adaptive Non-Maximum Suppression ensures temporal diversity. Evaluation on a custom dataset demonstrates tracking stability, achieving an IDF1 of 97.89% and a MOTA of 95.79%. Compared to Top-K selection, our algorithm also increases visual diversity by 146%, temporal coverage by 89%, and information retrievability by 3.5%.
[CV-76] PartLLM : A Unified Multimodal Foundation for 3D Part Segmentation SIGGRAPH
链接: https://arxiv.org/abs/2609.25832
作者: Zhe Zhu,Yiheng Zhang,Peng Li,Zixing Zhao,Honghua Chen,Yaqing Zhang,Le Wan,Zhiyang Dou,Cheng Lin,Yuan Liu,Mingqiang Wei,Wenping Wang
类目: Computer Vision and Pattern Recognition (cs.CV); Graphics (cs.GR)
备注: Accepted to SIGGRAPH Asia 2026 (ACM Transactions on Graphics). Project Page: this https URL
Abstract:Part segmentation is a fundamental problem in computer graphics and 3D vision. Recent works have expanded 3D part segmentation beyond fixed taxonomies, but existing approaches typically only address a specific setting, such as text-guided part segmentation or point-based interaction. In this work, we argue that these settings can be unified as an intent-conditioned generative problem, where different prompts specify the desired part decomposition. To this end, we introduce PartLLM, a unified multimodal model that formulates 3D part segmentation as autoregressive semantic decomposition. Conditioned on an input shape and a user prompt, PartLLM autoregressively generates semantic part hypotheses as queries for mask prediction and feeds them to a decomposition-aware decoder that jointly predicts coherent part masks. This unified design supports text-guided part segmentation, interactive segmentation, and full-shape semantic decomposition with controllable granularity within a single model. Extensive experiments across these task settings show that PartLLM consistently outperforms task-specific baselines, demonstrating the effectiveness of unifying 3D part segmentation under an intent-conditioned generative formulation.
[CV-77] Sometimes You Gotta Run Before You Can Walk: Run-then-Walk Scheduling Strategy for VLM Autonomous Driving
链接: https://arxiv.org/abs/2609.25831
作者: Yuqi Ye,Shangkun Sun,Junhong Lin,Jiayi Zhao,Changhao Peng,Wei Zheng,Guoqing Liu,Tiesong Zhao,Wei Gao
类目: Robotics (cs.RO); Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Recent VLM-based autonomous driving planners adopt GRPO-style reinforcement learning to optimize driving performance. However, existing GRPO recipes either optimize driving efficiency, risking progress-seeking but unsafe behavior, or enforce early safety constraints, leading to overly conservative behavior; both require lengthy training. To solve these problems, we first reveal two distinct RL regimes: a progress regime (Run-GRPO) that aggressively explores high progress, and a safety regime (Walk-GRPO) that restores safety under stable progress. Based on this finding, we propose \textitRun-then-Walk , a simple yet effective two-stage reward scheduling strategy for GRPO, achieving both better performance and faster convergence. Unlike one-stage RL, which may focus on progress, safety, or a mixture of both within a single training phase, this schedule explicitly separates progress discovery from safety repair. In the \textitRun phase, we focus on progress, allowing the policy to escape the conservative bias and discover high-progress modes. In the subsequent \textitWalk phase, we introduce endpoint and safety strategy to repair unsafe behaviors from the Run phase. This reversed schedule overcomes the conservatism of Walk-first methods and the unsafe progress-seeking of joint optimization. We validate it with various VLM-based planners on multiple benchmarks: NAVSIMv1, NAVSIMv2, Navhard, and nuScenes. Extensive experiments demonstrate improved driving performance while requiring 40–50% fewer RL training epochs than the baselines.
[CV-78] MorphoSHAP: Rethinking the Unit of Attribution in Explanation for Deep Visual Models
链接: https://arxiv.org/abs/2609.25815
作者: Anirudh Prabhakaran,Alexandre Rocchi,Gianni Franchi
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
备注: 21 pages
Abstract:Visual attribution methods typically explain predictions using pixels, superpixels, or regular patches. These representations can localize important regions, but provide limited information about their structure. We introduce MorphoSHAP, a model-agnostic post-hoc method that instead uses morphological shapes as the players of a Shapley attribution game. Using the Tree of Shapes, each shape is described by its scale, geometry, and signed contribution, providing explanations of where the evidence lies, what type of structure carries it, and how strongly it affects the prediction. This shared morphological vocabulary enables spatial, textual, and global class-level explanations beyond image-specific heatmaps. To the best of our knowledge, MorphoSHAP is the first SHAP-based image attribution framework to combine these different forms of explanation. Across five diverse datasets and three architectures, MorphoSHAP achieves strong insertion/deletion performance and outperforms competing attribution methods on several benchmarks. Finally, a user study shows that MorphoSHAP provides explanations that are easy to use and are preferred over standard attribution baselines.
[CV-79] LiFR v2: Completion-Augmented Event Propagation for High-Rate Dense Prediction
链接: https://arxiv.org/abs/2609.25803
作者: Tao Wan,Xiaoshan Wu,Yifei Yu,Bo Wang,Xiaoyang Lyu,Muxin Liu,Aoxuan Pan,Zhongrui Wang,Xiaojuan Qi
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: 15 pages, 9 figures, 6 tables
Abstract:High-rate dense perception in dynamic environments is limited by the low update rate of RGB cameras, as rapid scene changes can occur between frames. Event cameras offer temporally dense but spatially sparse measurements, complementary to spatially dense RGB observations. Direct fusion cannot fully exploit this complementarity, while event-guided propagation fails on newly appearing or disoccluded regions without valid RGB support. We present LiFR v2, a unified propagation-completion-memory framework for causal anytime and streaming dense prediction from an RGB keyframe and events. LiFR v2 introduces an Event-Guided Completion Module (EGCM) to recover task-relevant representations where propagation is unsupported, and a History Retrieval Module (HRM) to reuse completed representations across successive queries. The framework supports semantic segmentation, monocular depth estimation, and multi-task dense prediction, and we further introduce SHF-Emerge to evaluate rapid object emergence and disocclusion. LiFR v2 achieves 74.37% mIoU on DSEC and 56.13% on SHF-Emerge, improving LiFR-Seg by 1.85 percentage points on the latter, while reducing SHF-Emerge depth RMSE from 1.564 m to 1.118 m over the propagation baseline. It also exceeds 100 FPS for both segmentation and depth, demonstrating accurate and efficient high-rate perception beyond RGB frame rates.
[CV-80] When Point Clouds Outperform Pixels: Rethinking Zero-Shot Multimodal Anomaly Detection
链接: https://arxiv.org/abs/2609.25793
作者: Chenglin Ye,Lupeng Liu,Dongbo Yu,Jun Xiao,Yunbiao Wang
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Zero-shot multimodal anomaly detection commonly assumes that RGB and point cloud modalities are equally reliable and can contribute uniformly to anomaly localization. We challenge this assumption. Using a set of recently proposed stringent metrics that penalize false anomaly responses in normal regions, we find that point clouds are substantially more reliable than RGB under zero-shot category shift. Motivated by this observation, we propose WOOPS (\textbfWhen P\textbfoint Cl\textbfouds Out\textbfperform Pixel\textbfs), a reliability-aware zero-shot multimodal anomaly detection framework. To strengthen the more reliable geometric modality, we design a Multi-view Information Decoupling module to suppress heterogeneous information from multi-view point cloud projections and enhance point cloud feature quality. To avoid unconditional fusion, we further introduce a Modality Reliability Calibration module to adaptively calibrate modality contributions according to their reliability. Extensive experiments show that our method achieves the best or competitive performance under the new metrics in both unimodal and multimodal settings. Further analysis demonstrates that point cloud information also improves RGB-only inference, while ablations verify the effectiveness of both modules. Code will be released upon acceptance.
[CV-81] RACE: Trajectory Representation and Consistency Estimation for AI-Generated Video Detection
链接: https://arxiv.org/abs/2609.25775
作者: Huangsen Cao,Hongkang chu,Siyao Yu,Xin Ding,Jianfeng Dong,Yongwei Wang
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Recent advances in generative video models have enabled the synthesis of visually realistic content, posing significant challenges to synthetic video detection. Existing detectors often rely on appearance artifacts, semantic inconsistencies, and temporal patterns that may be generator-specific, limitating generalization to unseen synthesis models. We investigate whether responses to a pretrained generative model provide more transferable forensic cues. Our key observation is that real and AI-generated videos exhibit distinct \emphvelocity responses under a pretrained Flow Matching video model. This distinction persists when different pretrained video-generation backbones are used as probes, suggesting that velocity responses offer transferable forensic signals beyond visual artificts. Motivated by this observation, we propose \textbfTRACE (\emph\underlineTrajectory \underlineRepresentation \underlineand \underlineConsistency \underlineEstimation), a generation-process-aware framework for AI-generated video detection. TRACE leverages a pretrained video DiT as a velocity-field probe to extract representations at multiple flow time points, and models cross-frame consistency through velocity differences between adjacent frames. We further introduce a \emphReal-Centered Trajectory Optimization objective that encourages generator-invariant representation learning. Extensive experiments on AIGVDBench demonstrate that TRACE generalizes effectively across diverse generators, substantially outperforming prior state-of-the-art methods on unseen open- and closed-source video generation models.
[CV-82] Video-HopChain: Multi-Hop Questions and Confidence-Gated Exploration for Video Reasoning Models
链接: https://arxiv.org/abs/2609.25773
作者: Trung Nguyen Quang,Yuhao Dong,Shuo Sun,Shuai Liu,Shulin Tian,Kim-Hui Yap,Ziwei Liu
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
备注:
Abstract:HopChain has shown on still images that multi-hop data synthesis improves vision-language reasoning, because long chain-of-thought reasoning exposes errors that compound across steps, while most data used for reinforcement learning with verifiable rewards (RLVR) rarely demands a chain of visual evidence, so these weaknesses are likely to stay unexposed. We observe the same problem in video, where this framework has not yet been explored. We therefore build Video-HopChain, a dataset of 22,550 multi-hop video questions over 13,378 videos, together with a held-out benchmark of 1,000 questions. Each question chains three to six yes/no questions about moments in one video, and each yields one of two integers depending on its answer. The final answer is the sum of these integers, so an exact match on that sum gives the verifiable reward that RLVR needs. We first train Qwen3-VL-8B with GRPO on a standard video dataset, and a second stage on Video-HopChain then raises the mean over eight video understanding and reasoning benchmarks from 55.4 to 57.9 and improves every one of them. Training on such a dataset, however, exposes a known limitation of GRPO: its learning signal comes from the reward variance within a group, so hard questions whose rollouts are all incorrect and easy questions whose rollouts are all correct both leave the group with no gradient. To recover these groups at the same compute budget, we introduce Confidence-Gated Exploration (CGE). With 8 rollouts per question, CGE samples the first 4 as usual. If these 4 are either all correct or all incorrect, it samples the last 4 with the policy’s most confident token masked inside the reasoning span, and removes the masked positions from the loss while all 8 rollouts enter the advantage. With CGE, the mean rises further to 59.3. We release the dataset, the checkpoint, and the data generation and training code.
[CV-83] Reading Right Answering Wrong: How Visual Configuration Changes Affect Evidence Use in VLMs
链接: https://arxiv.org/abs/2609.25770
作者: Dingyang Lin,Yingfeng Luo,Chenglong Wang,Chenwei Zhu,Anxiang Ma,Jingbo Zhu,Tong Xiao
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: 6 pages, 3 figures, and 6 tables. Preprint
Abstract:Vision-language models (VLMs) have achieved strong performance on tasks such as visual question answering, yet small image resizes can turn correct answers into errors. We investigate whether changes in visual configuration, such as image tiling and token arrangement, contribute to this instability. Across seven checkpoints and four benchmarks, equally small resizes cause more correctness flips when they switch configurations. Surprisingly, in over half of these cases, models answer the question incorrectly but can still read the correct answer when told what to read. Furthermore, attention interventions in LLaVA-NeXT suggest that configuration changes can weaken the use of readable information during answering. We therefore guide models using field cues and their own transcriptions. With annotation assistance, these forms of guidance together correct 97.2% of errors with readable information. These findings show that configuration changes can affect how models use information they can still read.
[CV-84] Dual Covariance Gaussian Splatting SLAM: Decoupling Rendering and Registration for Robust Real-Time Tracking
链接: https://arxiv.org/abs/2609.25746
作者: Edward Beng Wai Tan,Siew-Kei Lam
类目: Robotics (cs.RO); Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:ICP-based 3D Gaussian Splatting (3DGS) SLAM tracks in real time by registering incoming frames against map Gaussians, using each primitive’s covariance for both rendering and registration. These two uses place conflicting demands on one covariance. The mapper shapes it to minimize photometric error, often flattening it against surfaces, while robust registration typically benefits from measurement uncertainty. We propose a dual-covariance parameterization. Each Gaussian keeps a single mean but holds two covariances: a rendering covariance optimized by the mapper, and a tracking covariance derived from an RGB-D sensor noise model. We further use the tracking covariances as Gaussian anchors for image corners, providing constraints in directions where depth geometry is weak. We evaluate on TUM RGB-D, ScanNet, Replica, and two outdoor sequences recorded with a RealSense D435i on wheeled and handheld platforms. We achieve robust tracking performance across multiple scenes and reduced odometry drift, while tracking at \sim 60 FPS.
[CV-85] SAMI3D-DW: Interactive Segmentation of Any 3D Medical Images
链接: https://arxiv.org/abs/2609.25743
作者: Ping Gong,Shiyuan Su,Fandong Zhang,Xinchen Han,Haowei Sun,Yiming Li,Yizhou Yu
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: 27 pages, 4 figures
Abstract:Interactive segmentation of 3D medical images supports quantitative analysis of anatomical structures and disease while allowing users to specify and refine their targets. Despite substantial progress by nnInteractive and VISTA3D, reliable segmentation across diverse clinical targets remains challenging, particularly for complex anatomical structures and the heterogeneous, long-tailed spectrum of pathology. We present SAMI3D-DW V1 (hereafter SAMI3D-DW), an interactive 3D segmentation model trained on Deepwise’s large-scale proprietary medical image datasets. We evaluate the model under simulated user interactions on a CT/MR benchmark comprising 4,326 cases from 219 source datasets, spanning 107 anatomical and pathological categories, organized by a medical taxonomy and evaluated with a category-balanced DSC score. SAMI3D-DW achieves the highest category-macro Dice among evaluated methods in both interaction modes. With one point, it scores 0.5764 versus 0.5315 for nnInteractive, the strongest baseline, rising to 0.7771 versus 0.7494 with five points. With bounding-box initialization, the scores are 0.7130 versus 0.6530. After five corrective clicks, SAMI3D-DW reaches 0.8002 versus 0.7868, making it the only evaluated box-compatible model to exceed 0.80. For radiologists and clinicians, SAMI3D-DW enables segmentation of complex anatomical structures, including intracranial vessel trees on CT and MR angiography, with a few clicks. In a preliminary in-house comparison involving neurofibromatosis type 1 (NF1), SAMI3D-DW-assisted tumor annotation took minutes per case and approximately one-fifteenth of the time required for manual annotation, highlighting its potential to support volumetric treatment-response assessment. Comments: 27 pages, 4 figures Subjects: Computer Vision and Pattern Recognition (cs.CV) Reportnumber: DW-AILAB-TR-2026-001 Cite as: arXiv:2609.25743 [cs.CV] (or arXiv:2609.25743v1 [cs.CV] for this version) https://doi.org/10.48550/arXiv.2609.25743 Focus to learn more arXiv-issued DOI via DataCite (pending registration) Submission history From: Ping Gong [view email] [v1] Tue, 22 Sep 2026 06:25:11 UTC (1,748 KB) Full-text links: Access Paper: View a PDF of the paper titled SAMI3D-DW: Interactive Segmentation of Any 3D Medical Images, by Ping Gong and 6 other authorsView PDFHTML (experimental)TeX Source view license Current browse context: cs.CV 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?) Author Venue Institution Topic About arXivLabs arXivLabs: experimental projects with community collaborators arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website. Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them. Have an idea for a project that will add value for arXiv’s community? Learn more about arXivLabs. Which authors of this paper are endorsers? | Disable MathJax (What is MathJax?) mathjaxToggle(); We gratefully acknowledge support from our major funders, member institutions, , and all contributors. About Help Contact Subscribe Copyright Privacy Accessibility Operational Status (opens in new tab) Major funding support from
[CV-86] Fysiverse-3D-Vision Technical Report: Generating Executable 3D Worlds from Images through Unified Spatial Reasoning
链接: https://arxiv.org/abs/2609.25741
作者: Dingkang Yang,Yizhou Liu,Wendong Cheng,Zizhi Chen,Shunli Wang,Yang Liu,Hongsheng Li,Lihua Zhang
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: Fysics AI Technical Report
Abstract:Generative models have advanced image-conditioned 3D content creation, yet generating controllable and executable 3D scenes from a single image remains challenging. Existing 3D generative approaches can synthesize visually plausible objects and scenes, but their spatial layout estimation is coupled with specific asset generators. They struggle to jointly model object semantics, metric geometry, and scene-level spatial relationships, which are essential for interactive editing, physical simulation, and embodied applications. We propose Fysiverse-3D-Vision, a unified vision-language-geometry framework for generative 3D scene reconstruction and executable asset construction from a single image. We establish a shared representation where spatial reasoning and geometric reconstruction mutually enhance each other, allowing object layouts to be inferred beyond the constraints of individual asset generators. Our model integrates textual supervision, semantic visual cues, and geometric representations within a unified Transformer to capture scene context, metric geometry, and object-level interactions. An object-conditioned layout module performs cross-attention between target object representations and global geometric features to predict object translation, rotation, and scale. Training progressively learns geometry-language alignment, introduces layout reasoning while preserving reconstruction capability, and refines physical consistency through collision-aware optimization. By separating spatial layout reasoning from asset synthesis, Fysiverse-3D-Vision provides an adaptable interface for interactive scene editing, object-level manipulations, and executable 3D content generation. Experiments demonstrate that our framework achieves superior geometric consistency, layout estimation, rendering quality, and physical property understanding compared with existing approaches.
[CV-87] Annual Earth-observation embeddings encode wildfire disturbance and support simplified burned area mapping
链接: https://arxiv.org/abs/2609.25731
作者: Jovana Knezevic,Clement Atzberger,Zhengpeng Feng,Adam F. A. Pellegrini,Srinivasan Keshav,David Coomes
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Medium-resolution (10-30 m) burned area mapping is vital for monitoring wildfires and their impacts, but remains difficult to scale. Existing methods require either curated fire-specific imagery or dense time-series analysis. Here, we tested whether annual Earth-observation embeddings retain wildfire disturbance signals sufficiently to map burned areas without either requirement. Using Tessera and AlphaEarth embeddings, we tested individual burn-scar delineation, mapping of all same-year fires within an area, regional wall-to-wall mapping, cross-continental transfer, and intra-annual fire timing. Tessera strongly encoded wildfire disturbance, allowing even linear models to separate burned from unburned pixels; the signal was weaker in AlphaEarth. Models trained on a single Tessera embedding matched or exceeded equivalent models using paired pre- and post-fire HLS imagery, and outperformed post-fire imagery alone. The same approach mapped all same-year fires within benchmark scenes (F1 = 0.90). Applied across California, with no California fire data used for downstream training, it recovered 97% of reference burned area and detected substantially more small and medium-sized fires than GABAM or MCD64A1. Separately, a model trained on 2018-2021 US fires transferred without retraining to 88 European fires from 2024-2025 (F1 = 0.88). For well-detected fires, ignition timing was recovered with a mean absolute error of 13 days. Performance declined for fires ignited near the end of the calendar year, and wall-to-wall deployment produced systematic false positives in some unseen landscapes. Annual embeddings nevertheless achieve high segmentation accuracy while moving the burden of dense time series processing upstream, providing a promising path towards simpler regional burned area mapping.
[CV-88] FoMo: Forking Moment in Generative Trajectory as a Perceptual Distance
链接: https://arxiv.org/abs/2609.25716
作者: Jaihyun Lew,Mingi Jung,Minjun Park,Wooseok Song,Sungroh Yoon
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Reference-based image quality assessment (IQA) metrics aim to reflect how humans perceive the perceptual distance between a pair of images. To learn how the human visual system (HVS) operates, recent reference-based IQA metrics heavily rely on human-annotated data. Mean opinion score (MOS)-based pointwise scoring, which assigns a scalar quality value per image, is preferable for annotation but is prohibitively expensive to collect at scale and is known to be noisy due to inconsistent human judgments. As an alternative, two-alternative forced choice (2AFC) pairwise labels have gained popularity due to their reliability and efficiency, but they capture only relative comparisons between pairs. In this paper, we propose a fully automated data generation pipeline that generates pointwise perceptual distance labels between image pairs without any human annotation. Our approach exploits the generative dynamics of diffusion models as a perceptual distance proxy, where the coarse structure of an image is generated in the early timesteps and the fine details are generated in the later timesteps. Images that fork early in the generation process share only coarse structure and are perceptually far apart; images that fork late differ only in fine detail. We demonstrate that the diffusion trajectory aligns well with the human visual system, and use this forking moment, FoMo, as a reference-grounded distance label to supervise the training of a reference-based IQA metric. The pointwise labels, which support universal comparison between arbitrary image pairs, enable an information-rich training objective. Extensive experiments across diverse backbone architectures confirm the effectiveness of our generation pipeline, outperforming human-annotated datasets in multiple benchmarks.
[CV-89] Interpretable AI plus Handheld Portable Retinal Photographs: A Low-Cost Glaucoma Screening Solution for West Africa
链接: https://arxiv.org/abs/2609.25697
作者: Charis Y. N. Chiang,Tarela Sarimiye,Adeyinka Ashaye,Martin Buist,Michael A. Hauser,Olusola Olawoye,Michaël J.A. Girard
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
备注: 31 pages, 2 Tables, 5 Figures, 1 Supplementary Material
Abstract:Purpose: To develop and evaluate an interpretable artificial intelligence (AI) framework for glaucoma screening from low-cost portable, handheld retinal fundus photographs in a West African population and to compare its performance with clinical tabletop fundus imaging. Methods: We used data from a community-based study of 681 participants (1,362 eyes) in Nigeria, comprising 414 glaucoma, 478 glaucoma suspect, and 470 non-glaucoma eyes. Fundus photographs were acquired using the low-cost handheld, portable Volk Viva retinal camera and the Canon CR-2-AF tabletop camera. We fine-tuned component models separately to each device to perform vessel segmentation, cup and disc boundary segmentation, and feature extraction to detect optic nerve head features. A final classification model combined these components to classify scans as glaucoma, glaucoma suspect or non-glaucoma. Feature-weight analysis and Gradient-weighted Class Activation Mapping were used for interpretation. Results: The models performed well on both Volk Viva and Canon CR-2-AF images: Vessel segmentation: 0.98 Dice Coefficient (DC) (Volk) and 0.94 DC (Canon); Cup and disc segmentation: 0.95 DC (Volk) and 0.96 DC (Canon); Optic nerve head feature detection: area under the receiver operating characteristic curve (AUCs) of 0.83 \pm 0.03 (Volk) and 0.87 \pm 0.04 (Canon); Classification model: AUCs of 0.85 \pm 0.01 (Volk) and 0.93 \pm 0.01 (Canon). Reports for each image, present model decision confidence scores and decision-rationale visualizations to support clinical interpretation. Conclusions: Volk Viva results were reasonably comparable to Canon CR-2-AF in the component models and not far behind in classification. This shows that interpretable AI combined with low-cost, portable imaging may enhance community-level glaucoma screening, especially in settings with limited specialist access and resources.
[CV-90] C2FXNet: Coarse-to-Fine Scene Expert for Unified Object Detection across Adverse Weather ACM-MM2026
链接: https://arxiv.org/abs/2609.25693
作者: Tianle Fang,Zhenbing Liu,Chong Yin,Bolun Li,Haoxiang Lu
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: 10 pages, 8 figures. Accepted at ACM Multimedia (ACM MM 2026)
Abstract:Object detection in adverse weather remains challenging because severe degradations weaken visual quality and disrupt semantic feature representations across diverse scenes. Existing methods usually rely on condition-specific designs, which limits their ability to generalize within a unified detector. In this paper, we propose a Coarse-to-Fine Scene Expert Network (C2FXNet) that achieves unified detection through hierarchical scene guidance. Specifically, C2FXNet introduces a dual-level guidance mechanism consisting of a Multi-step Reasoning Router (MRR), which performs GRU-based recurrent scene reasoning over compressed multi-scale visual cues and frozen coarse scene prototypes, and a Fine Scene Refinement (FSR) module, which uses image-specific semantic cues to modulate high-level features for local variation handling. Furthermore, a Scene-aware Mixture-of-Experts (SMoE) dynamically combines scene-specific experts under the joint guidance of MRR and FSR. By coupling coarse scene reasoning with fine-grained semantic refinement, C2FXNet enables robust multi-scene detection without scene-specific training. Extensive experiments on RTTS, ExDark, and our newly constructed Adverse Weather Dataset (AWD) demonstrate that C2FXNet consistently outperforms state-of-the-art methods across foggy, dark, and clear conditions, reaching 63.70%, 71.14%, and 54.19% mAP on RTTS, ExDark, and AWD, respectively. The source code will be released at this https URL.
[CV-91] Initialization and Stopping Tolerance in CPU Dermoscopic Segmentation
链接: https://arxiv.org/abs/2609.25685
作者: Wenhao Xu,Yixian Kong,Ting Pan,Changwei Wang,Feilong Wang,Rongtao Xu
类目: Computer Vision and Pattern Recognition (cs.CV); Image and Video Processing (eess.IV)
备注: 10 pages, 3 figures, 2 tables
Abstract:Contour initialization and numerical stopping can jointly affect the evaluation of active-contour segmentation. We examine their interaction using the open-source scikit-image Chan-Vese implementation on a resized ISIC 2017 mirror. A fixed development set of 100 images selects a common input channel; all 600 images in the repository’s held-out partition are then evaluated. Otsu thresholding is compared with checkerboard-, disk-, and Otsu-initialized contours under default and tighter level-set tolerances. At the default tolerance, Otsu initialization increases mean image Dice from 0.6011 to 0.6660 relative to checkerboard initialization, a paired difference of 0.0649 (95% image-bootstrap interval [0.0452, 0.0860]). Otsu thresholding alone achieves 0.6897. The default disk initializer stops after one iteration on 471 images. Tightening the tolerance reduces the Otsu-seed advantage over checkerboard initialization to 0.0197, with most runs reaching the 500-iteration limit. The default-tolerance advantage also reverses between small- and large-lesion strata. These findings show that an improvement over a generic initializer can coexist with deterioration relative to the threshold baseline. Evaluations should retain the unrefined mask as a comparator and report the initial-field definition, stopping tolerance, and observed iteration counts together.
[CV-92] Real-Time Atomic-Resolution Electron Phase Imaging without Probe Calibration via Ptychography-Supervised Learning
链接: https://arxiv.org/abs/2609.25684
作者: H. Yue,C.-C. Chen,C.-N. Hsiao,J. Cheng,Y. Liu,X.Z. Liao,Steve F. Shu
类目: Computer Vision and Pattern Recognition (cs.CV); Materials Science (cond-mat.mtrl-sci)
备注: Submitted to Advanced Science on June 18, 2026
Abstract:Atomic-scale phase imaging is central to resolving defects, interfaces, and weakly scattering atoms that govern the behavior of nanoscale materials. Electron ptychography delivers sub-ångström phase sensitivity but remains an offline technique, because its iterative reconstruction is computationally expensive and sensitive to experimental calibration, preventing live use during data acquisition. Here, a ptychography-supervised local inference framework is presented that converts four-dimensional scanning transmission electron microscopy (4D-STEM) into an acquisition-compatible phase-imaging workflow. Physics-constrained reference phase maps reconstructed from a single experimental AuPd dataset serve as teacher labels for a compact model that predicts local phase patches directly from diffraction measurements, without explicit probe input or online iterative optimization. Full-field images are assembled by deterministic overlap stitching. The workflow reaches an online latency of about 0.27 ms per probe position and a throughput of about 20,000 positions per second, an approximately 1,000-fold speed-up over GPU-accelerated ePIE, while preserving atomic-scale lattice contrast and reciprocal-space fidelity. Without fine-tuning, the same model transfers across materials (WS2), defocus conditions (high-entropy alloy nanoparticles), and instruments (hBN at 300 kV). The approach amortizes ptychographic redundancy into a fast, generalizable workflow that enables real-time atomic-scale phase imaging for materials microscopy.
[CV-93] CODA: Depth-Aligned Scene Completion and Object Decomposition from a Single RGB-D Image
链接: https://arxiv.org/abs/2609.25654
作者: Dongwon Son,Junhyek Han,Yoontae Cho,Minseok Lee,Hong-seok Choi,Jiwook Choi,Hyungjin Kim,Beomjoon Kim
类目: Robotics (cs.RO); Computer Vision and Pattern Recognition (cs.CV); Machine Learning (cs.LG)
备注: 8 pages, 7 figures, 3 tables. Project page: this https URL
Abstract:Robots operating safely in cluttered everyday environments often need to infer scene geometry from partial observations. Methods that detect objects in 2D and reconstruct them independently struggle in such scenes: a missed object is never reconstructed, a merged detection can fuse two objects, and separately reconstructed meshes may overlap or fail to touch their supporting surfaces. We introduce CODA (Complete Once, Decompose Afterward), a generative model that instead reconstructs the complete scene geometry from a single unsegmented RGB-D image, then separates the surface into the surrounding environment and movable objects. Still, generated scene geometry can drift from the observed partial point cloud. To reduce this drift, CODA uses two explicit 3D grounding mechanisms to keep reconstructed geometry consistent with observed surfaces while completing unseen regions. Experiments on HomebrewedDB and our custom cluttered-scene dataset show more accurate reconstructions and a higher fraction of objects remaining in place under simulated gravity than both object-first and scene-first baselines.
[CV-94] GameDirector: Decoupling Gameplay Logic from Rendering for Player-Configurable Game World Models
链接: https://arxiv.org/abs/2609.25652
作者: Zijun Lin,Zhiyang Deng,Yuzhe Wu,Bihan Wen,Yeying Jin
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: Project Page: this https URL
Abstract:Recent game world models support realistic visual simulation and interactive gameplay based on player inputs. However, they typically learn environment dynamics from pixel-level supervision, jointly modeling perception, memory, state transitions, and rendering within a single end-to-end framework. While this design enables open-ended, action-controllable generation, it still falls short of delivering a complete gameplay experience. Games are governed by explicit mechanics, such as health deduction, skill activation, combat rules, and termination conditions. These mechanics depend on precise and consistent state transitions that generative models alone cannot reliably enforce. In contrast, game engines can guarantee such mechanics through hard-coded rules, but provide limited flexibility for player-driven creation. To bridge these paradigms, we introduce GameDirector, the first agentic framework that decouples rule-based gameplay logic from visual rendering. Given player-defined configurations, the framework acts as an intelligent director that interprets visual observations, updates game states, tactically controls NPCs, and enforces gameplay rules. It then translates these decisions into text prompts that guide the video world model to render the resulting gameplay. This separation allows players to configure characters, states, and rules much like a game developer while preserving coherent game mechanics. Experiments on three games, using data collected by our automated gameplay agent, show that GameDirector achieves accurate state tracking, reliable rule following, and improves boss action quality by more than 39.9% over various end-to-end game world model settings. Overall, by externalizing player-controllable game logic, GameDirector establishes a middle ground between hard-coded simulation and generative modeling, enabling more flexible and closed-loop gameplay experiences.
[CV-95] Decoupling Disease Covariates and Individual Variability: A Unified Disentanglement Framework for Medical Image Classification
链接: https://arxiv.org/abs/2609.25650
作者: Shengjie Zhang,Jinglin Zhang,Zhuangzhuang Jiang,Ziqi Yu,Yipin Zhang,Qi Zhang,Xiang Chen,Haibo Yang,Fei Gao,Longbiao Cui,Yuan Zhou,Xiao-Yong Zhang,Alzheimer’s Disease Neuroimaging Initiative
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: 14 pages, including a 4-page appendix
Abstract:Accurately isolating disease-related features from confounding covariates (e.g., age, gender, site) and individual variations remains a fundamental challenge in medical image classification. Traditional regression-based approaches may ignore non-linear relations between image features and true covariates. To overcome this issue, we present a generalized Medical Imaging Disentanglement Learning (MedIDL) framework. MedIDL maps image features into three mutually orthogonal latent spaces through specialized disentanglement heads: a disease classification head guided by a supervised loss, a covariate-alignment head constrained by cross-subject similarity matching, and a Gaussian head absorbing individual variations. We evaluated our framework across 7 datasets encompassing diverse imaging modalities. MedIDL outperforms state-of-the-art supervised and self-supervised classification methods in accuracy across all datasets. Association analyses demonstrate that MedIDL successfully isolates target-specific latent representations. Gradient-based interpretability mappings localize pathognomonic patterns aligning with established clinical literature.
[CV-96] What Drives Hierarchy-Aware Image Retrieval? Taxonomy Alignment Objective Choice and Geometry
链接: https://arxiv.org/abs/2609.25638
作者: Ling Shi(Southeast University)
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: 17 pages total: 9-page main paper (including references) + 8-page supplementary material; 3 figures and 2 main-paper tables
Abstract:Foundation vision models provide strong generic representations, yet high class-level retrieval accuracy does not necessarily imply that an embedding respects a target semantic taxonomy. We study strict explicit-taxonomy image retrieval on frozen DINOv2 features and ask: when hierarchical retrieval improves, how much of the change is associated with the organization of taxonomy-aware supervision, and how much with the Euclidean-hyperbolic geometry choice? We evaluate higher levels with strict cross-class criteria that exclude finer-grained matches, and compare Euclidean and hyperbolic projections trained with taxonomy-distance regression or a taxonomy-aware supervised contrastive objective. A compute-matched 2 x 2 Geometry x Loss factorial uses the same 768-256-32 projector capacity, optimization schedule, batch order, and fixed 100-epoch budget; the Loss axis denotes the Regression-to-Taxonomy-SupCon objective-family contrast. On CUB, the objective-family contrasts in mean hierarchy mAP (strict middle/high average, excluding Class/Leaf) are +0.0487 in Euclidean space and +0.0414 in hyperbolic space, compared with geometry contrasts of +0.0102 and +0.0030. On NABirds Parent-disjoint retrieval, the corresponding objective-family contrasts are +0.0467 and +0.0440, whereas geometry contrasts are +0.0017 and -0.0009. A semantic-alignment control shows that the true taxonomy substantially outperforms a structure-preserving shuffled hierarchy, while a NABirds curvature/radius control does not support stronger negative curvature as the explanation for the observed hierarchy gains. Across the two taxonomies, the Regression-to-Taxonomy-SupCon contrasts are larger in aggregate than the evaluated geometry contrasts; semantic alignment also matters separately, while geometry remains hierarchy-dependent. Comments: 17 pages total: 9-page main paper (including references) + 8-page supplementary material; 3 figures and 2 main-paper tables Subjects: Computer Vision and Pattern Recognition (cs.CV) Cite as: arXiv:2609.25638 [cs.CV] (or arXiv:2609.25638v1 [cs.CV] for this version) https://doi.org/10.48550/arXiv.2609.25638 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
[CV-97] Shallow to Deep: Aligning Token Pruning with Stage-wise Roles in LVLMs EMNLP2026
链接: https://arxiv.org/abs/2609.25635
作者: Shuo Zhang,Jintao Tong,Yixiong Zou,Yuhua Li,Ruixuan Li
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: Accepted to EMNLP 2026. 17 pages, 10 figures, 10 tables
Abstract:Large Vision-Language Models (LVLMs) incur high computational costs from redundant visual tokens. Although training-free attention-based multi-layer pruning in the vision encoder stage has been explored as an effective strategy, we find that pruning in shallow layers consistently degrades performance. In this paper, we aim to understand this problem and seek a solution. By analyzing attention patterns across network depth, we find that shallow layers primarily function as edge detectors with chaotic attention maps, while deeper layers transition through local subject recognition and unstable semantic aggregation. To address the misalignment between pruning strategies and network stages, we propose STD, a hierarchical token pruning framework that adapts token selection mechanisms to the functional role of each network stage. STD employs High-Frequency Spectral Analysis in shallow layers to deterministically preserve structural edges, uses Gaussian-Smoothed Attention in intermediate layers to maintain spatial coherence, and introduces a Stability-Adaptive Trigger in deep layers to execute pruning only during semantically stable phases. Extensive experiments show that STD outperforms state-of-the-art pruning methods by 1.1% on LLaVA-1.5-7B with 88.9% token reduction, while also being plug-and-play and highly effective when combined with other methods, and by 2.1% on LLaVA-NeXT-7B with 94.4% reduction, delivering a 3.9x speed-up in the prefilling stage. Our code will be released at this https URL.
[CV-98] MachEmbodied-U0: Unified Understanding and Generation Model for Embodied Intelligence
链接: https://arxiv.org/abs/2609.25627
作者: Haoran Wen,Wenfu Wang,Kunsong Shi,Jingke Wang,Wancheng Feng,Yiren Zhang,Yueran Zhao,Xuancheng Zhang,Nanfei Ye,Xingru Chen,Zhaohong Sun,Chengmin Yang,Zikang Yu,Penghao Bi,Jia Shi,Yu Liu,Kun Zhan,Yan Xie
类目: Robotics (cs.RO); Computer Vision and Pattern Recognition (cs.CV)
备注: Technical report. Project page: this https URL . Code: this https URL
Abstract:General-purpose robot control requires models to understand task intent, identify where to interact, capture how the scene evolves, and generate precise actions. Vision-language-action models provide strong semantic priors but typically do not explicitly model scene dynamics, while world-action models couple visual prediction with control without necessarily exposing the task-relevant semantic and spatial structure needed for fine-grained manipulation. We present MachEmbodied-U0 (ME-U0), a unified embodied foundation model connecting understanding and generation experts through a Mixture-of-Transformers architecture. Subtask prediction and affordance grounding guide joint visual-dynamics and action generation via flow matching. Visual dynamics encompass future RGB, depth, surface normals, and optical flow, providing complementary supervision for appearance, geometry, and motion. Multi-rate Rotary Position Encoding (MRPE) aligns visual dynamics with fine-grained control. We pretrain ME-U0 on approximately 4,200 hours of curated demonstrations from robotic datasets and egocentric datasets. Using only the supervision natively available in each downstream benchmark, ME-U0 achieves an average score of 17.66 on the RoboDojo simulation benchmark and average success rates of 99.0% and 82.5% on LIBERO and LIBERO-Plus, respectively. We additionally validate ME-U0 on real-world robotic manipulation tasks, demonstrating its effectiveness beyond simulation. Without corresponding downstream supervision, ME-U0 further demonstrates zero-shot subtask prediction, affordance grounding, and visual dynamics on simulated and real-world observations. Overall, ME-U0 combines competitive downstream control performance with transferable task-grounding and visual-dynamics capabilities across simulation and the real world.
[CV-99] Evidence-gated multimodal parsing and vectorization of architectural floor plans
链接: https://arxiv.org/abs/2609.25615
作者: Hongxuan Chen,Wenda Wang,Jiachen Lu,Qirui Shen,Zilong Huang,Lei He,Xinyue Dong,Weixin Huang
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: 33 pages, 43 figures, 27 tables
Abstract:Architectural floor plans remain a high-friction barrier to archive digitization and early design-model preparation because heterogeneous graphics encode spatial semantics and editable geometry together. We introduce SALI-FP, an evidence-gated multimodal pipeline that converts a plan into reviewable semantic maps, objects, vectors, and relation records while constraining local revisions by image evidence. In a full production audit of 11,534 heterogeneous plans, SALI-FP produced structured outputs for every plan, including 752,510 valid polygon-bearing objects. The same output form has supported initial drawing digitization and design-model preparation in practical design work. Public-benchmark calibration is paired with a 30-case matched visual evidence set in Appendix F, where room-scale coverage, openings, oblique boundaries, and circulation continuity can be inspected directly. SALI-FP offers an engineering-oriented interpretation-to-geometry workflow for reviewed CAD/BIM preparation and existing-building information recovery.
[CV-100] Ultra-fast Neural Inference for Stochastic Gaussian Splatting Denoising WWW
链接: https://arxiv.org/abs/2609.25604
作者: Chenxiao Hu,Hao Zhang,Yanchen Zhang,Meng Gai,Guoping Wang,Sheng Li
类目: Computer Vision and Pattern Recognition (cs.CV); Graphics (cs.GR)
备注: Video supplements: this https URL this https URL
Abstract:Stochastic rendering eliminates the sorting and alpha blending process in Gaussian splatting, at the cost of introducing spatial noise. Formulating temporal denoising over the pixel stream shared by view-consistent stochastic splatting renderers, we propose a temporal neural denoiser validated on stochastic 2D Gaussian Splatting rendering, combining dual-path exponential moving average accumulation, per-pixel learned trust prediction for history validation, a fixed anisotropic spatial filter and a variance-gated composition with stabilization. The denoiser suppresses the noise, achieving temporally stable, visually compelling outputs during free camera navigation, all while retaining the sort-free, blend-free rasterization performance. The combined pipeline retains a PSNR gap to sorted alpha-blending renderers, but the denoiser’s overhead stays below the time saved by removing sorting and blending.
[CV-101] Observer Choice and Threshold Selection in Retinal Vessel Segmentation: A Subject-Separated Evaluation
链接: https://arxiv.org/abs/2609.25597
作者: Wenhao Xu,Yixian Kong,Ting Pan,Changwei Wang,Feilong Wang,Rongtao Xu
类目: Computer Vision and Pattern Recognition (cs.CV); Image and Video Processing (eess.IV)
备注: 7 pages, 3 figures
Abstract:The annotation used to select a segmentation threshold is part of the evaluation protocol, yet its effect is easily conflated with model quality. We examine this choice for retinal vessel segmentation using all 28 CHASE DB1 images and both human annotations. A fixed seven-fold protocol keeps both eyes of each of the 14 subjects together. Random forests and Extra Trees are fitted against observer 1 with three random seeds, yielding 42 fits. Five threshold policies share identical score maps: fixed 0.50, observer-1 tuning, observer-2 tuning, mean-observer tuning, and maximin tuning of the per-image lower observer Dice. For random forests, maximin changes the threshold in 19 of 21 fits, but worst-observer Dice decreases from 70.53 percent to 70.45 percent. The paired difference is -0.073 percentage points, with a conditional subject-bootstrap 95 percent interval of [-0.384, 0.238]. Extra Trees shows the same direction. Identical observer-1-tuned random-forest masks score 73.66 percent against observer 1 and 71.06 percent against observer 2. The results support explicit reporting of both the threshold-selection reference and evaluation reference; they do not support an accuracy benefit from maximin tuning in this cohort. All splits, raw predictions, metrics and code are supplied. AI assistance is disclosed.
[CV-102] Hi-OPD: Hierarchy-Aware Open-Prompt Detection for Remote Sensing Images
链接: https://arxiv.org/abs/2609.25584
作者: Jinlong Hu,Yi Zhang,Zhiqi Xia,Yikang Zhou,Shunping Ji
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Hi-OPD addresses a failure mode left uncontrolled by flat open-prompt training: descendant retrieval need not persist under ancestor queries when multi-source remote sensing annotations exhibit inconsistent granularity and missing labels. A detector may localize \textitcar and \textitvan under atomic prompts yet miss the same instances under \textitvehicle; flat AP does not expose this cross-level inconsistency. We propose Hi-OPD, a hierarchy-aware open-prompt detector, and construct RS153-HierOPD from 175,644 retained training image/tile records and 3.48M boxes mapped to 153 atomic categories with sparse hierarchy and alias relations. Hi-OPD learns ancestor retrieval through hierarchy-safe negative sampling, path multi-positive supervision, and one-way upward consistency, while per-source risk exclusion handles potentially missing labels. ConvVPE converts K-shot support boxes into text-compatible embeddings using detector-native features and the shared contrastive head. On Track A, Hi-OPD obtains 79.7/72.3 AP50 on DIOR/DOTA-v2.0, above the literature-reported OpenRSD results of 76.7/71.8. Under controlled training on the original converted annotations, the full hierarchy recipe raises DOTA-v2.0 parent AP50 from 7.2 to 71.5 and FAIR1M grandparent AP50 from 31.6 to 71.4, while DOTA-v2.0 atomic AP50 changes from 71.4 to 72.3. The text path reaches 99.7% CAR50 (0.3% violation) across the three common sources and 99.9%/0.1% on FAIR1M grandparent relations. On held-out VEDAI, text AP50 is 75.9, 6.2 points above OpenRSD. Joint AP and CAR show that explicit hierarchy training repairs this failure mode while retaining atomic detection and prompt transfer. Subjects: Computer Vision and Pattern Recognition (cs.CV) Cite as: arXiv:2609.25584 [cs.CV] (or arXiv:2609.25584v1 [cs.CV] for this version) https://doi.org/10.48550/arXiv.2609.25584 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
[CV-103] Agent ic Building-Aware Satellite Gaussian Splatting for Auditable Urban DSM Reconstruction
链接: https://arxiv.org/abs/2609.25578
作者: Wentao Sun,Zhengsen Xu,Yiping Chen,John S. Zelek,Jonathan Li
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: 7 pages, 6 figures
Abstract:Urban-scale 3D reconstruction from satellite imagery supports disaster response, city monitoring, and geospatial digital twins, yet neural rendering methods typically optimize average visual fidelity rather than the structures that analysts inspect first: buildings. We present an agentic building-aware satellite Gaussian Splatting workflow that uses Segment Anything-derived building masks as semantic priors and an Agentic Reconstruction Controller to select, verify, and record DSM reconstruction policies. On the DFC2019 JAX_004 scene, building-aware weighting reduces building-region DSM MAE from 0.844 m to 0.806 m, showing that semantic priors can shift reconstruction capacity toward analyst-critical regions. A staged schedule provides a balanced operating point, improving full-scene MAE from 1.362 m to 1.349 m while retaining a building gain. Across four JAX scenes, the Agent selects validated policies for both general DSM and building-focused DSM objectives, and produces building-inventory metadata and per-scene decision records. The system combines semantic priors, policy selection, region-specific DSM metrics, and DSM-derived GIS surface products for auditable urban 3D analysis.
[CV-104] RootQuantV2: Adapting a Vision Foundation Model for Root-Trait Regression from Minirhizotron Imagery ECCV2026
链接: https://arxiv.org/abs/2609.25567
作者: Kinjalk Parth,Sebastian Varela,Andrew D. B. Leakey
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI); Machine Learning (cs.LG); Quantitative Methods (q-bio.QM)
备注: 20 pages (15 main + 5 references), 4 figures, 5 tables. Accepted to the Computer Vision in Plant Phenotyping and Agriculture (CVPPA) Workshop at ECCV 2026. Code and weights: this https URL
Abstract:A lack of high-throughput phenotyping solutions for root traits in field-grown crops has severely constrained understanding and improvement of below-ground traits and processes. Minirhizotrons are the standard non-destructive root-phenotyping method in field environments. Computer vision solutions are needed to allow automated trait estimation at scale, but training data is scarce and human annotations are often inaccessible because they reside in proprietary software that only exports per-image scalar totals of root length and surface area. Nevertheless, large numeric archives of these root traits already exist. RootQuant showed that the traits can be predicted directly from the whole image by regression, thus removing manually traced masks from the pipeline; RootQuantV2 takes that idea further by replacing RootQuant’s CNN backbone with a self-supervised ViT. We adapt a frozen DINOv3 ViT-L/16 with a hybrid parameter-efficient scheme. Training only 11.9M parameters (3.78% of the model), RootQuantV2 achieves length and area R^2 of 0.950 and 0.930, respectively, while lowering length/area RMSE by 24.3%/20.7% over RootQuant. RootQuantV2 thus repurposes legacy numeric archives for high-throughput, automated root trait estimation.
[CV-105] Point Diffusion Mamba: Unified Diffusion-State-Space Modeling for Single-View 3D Reconstruction under Data Scarcity
链接: https://arxiv.org/abs/2609.25538
作者: Wei Zhou,Xinzhe Shi,Xingxing Hao,Xing Hao,Kang Li,Jinye Peng,Ying He
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:While single-view 3D reconstruction has seen significant progress, extrapolating complex 3D structures from inherently ambiguous 2D observations remains fundamentally ill-posed, particularly in the critically underexplored data-scarce regime. To address this challenge, we propose Point Diffusion Mamba (PDM), a method that integrates the generative power of diffusion models with the efficiency of state-space model for single-view 3D reconstruction under data-scarce conditions. Specifically, PDM employs a lightweight reconstruction module tailored to handle unordered point-cloud inputs effectively. By combining a Local Geometric Aggregation module with Mamba blocks, our approach jointly models global geometric structures and local details. In 3D reconstruction, each point in the initial noisy input requires a precise prediction, yet the high-level features extracted by the Mamba module capture only abstract semantic information from sparse points. To bridge this gap, we introduce the Hierarchical Feature Integration Network, which fuses high-level semantic and local geometric features for each point, overcoming the limitations of token-based point-cloud reconstruction. Furthermore, we propose a Dynamic Weighted Sampling strategy that adaptively unifies 3D generation with single-view reconstruction by leveraging generative priors to enhance reconstruction quality. Experimental results on the ShapeNet and Pix3D benchmarks demonstrate that PDM outperforms state-of-the-art methods, providing an effective solution for 3D reconstruction under data-scarce settings. Code is available at: this https URL.
[CV-106] Real-World Perception for Autonomous Driving in Adverse Weather: Enhancing Standard Detectors via Foundation-Guided Auto-Annotation
链接: https://arxiv.org/abs/2609.25515
作者: Sepideh Gohari,Goodarz Mehr,Azim Eskandarian
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: Submitted to IEEE for review
Abstract:Standard deployment-ready object detectors for autonomous vehicles degrade in adverse weather and lighting conditions without being trained on extensive domain-specific data. While large-scale vision foundation models offer robust zero-shot generalization, their high computational cost makes them impractical for real-time deployment. To bridge this gap, we propose a foundation-guided auto-annotation pipeline that enhances standard detectors without architectural changes. We first benchmark three distinct models, YOLOv8, Co-DETR, and SAM3, on our custom real-world driving dataset spanning 25 unique operational scenarios across various route, weather, and lighting conditions. Based on our analysis, SAM3 demonstrates superior accuracy and resilience across all scenarios. Thus, we deploy it as an offline auto-annotator to generate pseudo-labels on the unannotated subset of our dataset. Fine-tuning the baseline YOLOv8 on these annotations yields a 16.04% higher overall mean Average Precision (mAP) and improves cross-environmental stability compared to the baseline model, highlighted by a 32.73% and 28.65% mAP increase in Residential Direct Sunlight and Highway Fog, respectively. These results demonstrate that standard detectors can achieve environmental resilience without the need for extensive manual annotation or architectural modifications.
[CV-107] A Deployment Study of Identity-Gated Drone Gesture Control
链接: https://arxiv.org/abs/2609.25511
作者: Diyari Mohammed Salih,Ilyes Chaabeni,Naima Ait Oufroukh
类目: Robotics (cs.RO); Computer Vision and Pattern Recognition (cs.CV)
备注: 7 pages, 4 figures, 4 tables
Abstract:Vision-based gesture control accepts commands from any hand in the camera field of view, which is unsafe in shared indoor spaces. This paper presents IGate, an identity-gated control stack that includes gesture control and face tracking, in which commands are admitted only when an enrolled operator is verified. The system performs few-shot user enrolment from 20 initial face frames, without prior user-specific training: verification compares an embedding of the current face crop against the enrolled template by cosine similarity, while face tracking uses proportional correction. Gesture control is achieved by classifying extracted hand landmarks using an RBF-SVM trained on a custom dataset. Additionally, a hierarchical finite-state machine handles mode selection, default, and fallback behaviours. The approach is tested on a DJI Tello EDU, each component evaluated offline and in-flight across 270 trials (149 flown). Face verification yields a 0.32% offline equal error rate versus 19.3% in-flight. Under hover-locked conditions, the RBF-SVM gesture classifier outperforms the geometric rule (0.850 vs. 0.651 accuracy), with 82% of this gap stemming from the depth channel. All logs and reproduction scripts will be released.
[CV-108] SBMVTrack: Spike-Budgeted Multi-View Learning for Energy-Efficient UAV Tracking
链接: https://arxiv.org/abs/2609.25503
作者: Pengzhi Zhong,Jiwei Mo,Haolun Li,Ge Zheng,Jingqi Wang,Xinyi Bo,Shuiwang Li
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:With sparse and event-driven computation, spiking neural networks show great potential for achieving accurate and energy-efficient UAV visual tracking. However, existing SNN-based trackers typically use spike firing rates only for energy evaluation and lack explicit optimization of actual spike activity. To address this, we propose SBMVTrack, a fully spiking framework for energy-efficient UAV tracking. SBMVTrack introduces Energy-Weighted Spike Budgeting (EWSB). EWSB weights actual spike activity according to the computational cost of each spiking layer. It constrains the energy-weighted firing rate and saturation activity, thereby reducing redundant spike computations. To improve tracking performance under the spike budget constraint, we propose Masked Multi-View Target Modeling (MVTM). This method treats the initial template, online template, and search region from the same sequence as correlated temporal views. It enhances the robustness of target representations through cross-view feature completion and identity-consistency learning. Extensive experiments on multiple benchmarks demonstrate that SBMVTrack effectively reduces the average spike firing rate and theoretical energy consumption. Meanwhile, it maintains competitive tracking performance, achieving a better accuracy-energy trade-off. The source code will be released upon acceptance.
[CV-109] mbariml: a curation pipeline for turning deep-sea imagery and video into object-detection training data
链接: https://arxiv.org/abs/2609.25500
作者: Lonny Lundsten,Kevin Barnard,Dave Caress
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Training data quantity and quality greatly affect object detection model performance, regardless of model architecture. When using object detection models on video and images from the deep sea, in which the objects of interest, primarily organisms, are sparse, faint, and hard to identify, incremental improvements to object detector performance may require an iterative approach to data labeling and management. This paper presents mbariml, a python-based video and image analysis pipeline built around the data labeling management process. mbariml uses an Ultralytics YOLO detection model, runs it over still images or video, stores every detection as a reviewable region of interest, groups those regions by visual similarity so that a human can accept or reject them in bulk, and exports the result as training data, statistics, image sidecars, and additional metadata. The human review stage is the centre of the design: an annotator can validate, relabel, resize, delete, and draw entirely new localizations, and every one of those edits is written back to the same database the detector wrote to. Video receives particular attention: the software treats each tracker-produced track as a provisional observation and selects one representative frame instead of retaining every detection in the track. We describe the pipeline stage by stage, including the operational middle-third heuristic used for track observation selection.
[CV-110] RGSQ: Riemannian Geometry-Sensitive Quantization for Large Vision-Language Models
链接: https://arxiv.org/abs/2609.25492
作者: Zhiping Wu,Dongdong Ren,Yangchengyu Zhou,Zhengjie Zhang,Wenbin Li,Hongbing Pan,Yang Gao
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
备注:
Abstract:Large vision-language models (VLMs) can be efficiently deployed under stringent memory and latency constraints through post training quantization (PTQ). However, most PTQ methods are designed for unimodal large language models (LLMs). These methods treat quantization errors as isotropic perturbations under the Euclidean assumption, which provides weak guidance on directions most sensitive to quantization in VLMs. Consequently, directly adapting unimodal PTQ approaches or solely employing modality-specific scaling often leads to uneven bit-width distribution and inconsistent performance in low-bit settings. To address these challenges, we propose Riemannian Geometry-Sensitive Quantization (RGSQ), which formulates quantization as a reconstruction problem under a unified Fisher-Riemannian metric. RGSQ identifies modality-specific sensitive directions via Riemannian manifold mappings built from modality-partitioned empirical Fisher factors and fused into a modality-aware Kronecker-structured metric. We then apply geometry-aligned rotations to reorient the local tangent frame, steering low-bit perturbations toward loss-insensitive axes. Finally, we apply a whitening transformation that maps the Riemannian objective to an equivalent Euclidean form, enabling standard unimodal PTQ methods to evaluate multimodal quantization error under their original assumptions. Across an extensive and diverse set of mainstream VLM benchmarks, RGSQ achieves the highest accuracy and stability under extremely low-bit settings (W2A8 and W3A8). It outperforms VLM-aware baselines, such as MBQ and MQuant, by up to 5.9% and surpasses single-modality improvements by up to 8.6%.
[CV-111] SAM-V: Geometry-Aware Segment Anything for Multi-View Instance Segmentation
链接: https://arxiv.org/abs/2609.25490
作者: Jiangshan Gong,Yuqun Wu,Qiqian Fu,Yao Xiao,Chuhang Zou,Shenlong Wang,Derek Hoiem
类目: Computer Vision and Pattern Recognition (cs.CV); Machine Learning (cs.LG)
备注:
Abstract:Consistent multi-view object segmentation is critical for 3D perception and robotics, yet remains challenging under severe viewpoint and occlusion changes. Existing methods typically perform 3D instance segmentation on point clouds or rely on offline 2D mask-matching pipelines. However, 3D instance segmentation is limited by scarce 3D annotations, while offline 2D matching suffers from object identity ambiguity across frames. To leverage strong 2D and 3D priors jointly, we propose SAM-V (Geometry-Aware Segment Anything for Multi-View Instance Segmentation). Instead of combining the two priors through post-hoc matching, SAM-V directly integrates features from a feed-forward geometry model (VGGT) into a 2D segmentation foundation model (SAM), trained end-to-end for cross-view instance prediction. SAM-V introduces a prompt-fusion mechanism that enriches sparse SAM prompt tokens with view-specific camera tokens and local VGGT features, making the prompt representation both view-aware and spatially grounded, together with a mask decoder that attends to dense 2D and 3D features. By conditioning the mask decoding directly on multi-view geometry, SAM-V produces consistent multi-view segmentation of a prompted object in a single forward pass without offline mask matching or explicit 3D reconstruction. On the IGGT 3D tracking benchmark, where consistent instance identity across frames directly determines performance, SAM-V improves overall IoU by 5 points and frame-level recall by 12 points on the ScanNet++ split over the state-of-the-art multi-view instance segmentation baseline and leads on all metrics in the zero-shot ScanNet split. Our code and pretrained models are available at this https URL.
[CV-112] MIND the Gap: A Geographic Implicit Neural Representation with Adjustable Spatial Scale
链接: https://arxiv.org/abs/2609.25454
作者: Isaac Corley,Arjun Rao,Esther Rolf,Konstantin Klemmer,Evan Shelhamer,Nils Lehmann,Marc Rußwurm,Gengchen Mai,Nathan Jacobs,Hannah Kerner
类目: Computer Vision and Pattern Recognition (cs.CV); Machine Learning (cs.LG)
备注: this https URL
Abstract:Geographic measurements are often sparse, leaving large areas without labels for the quantities we want to map. Geographic implicit neural representations (INRs) address this by learning smooth, general-purpose embeddings that can be queried at any coordinate. Downstream models combine these embeddings with sparse labels to predict target values at unsampled locations without satellite imagery at inference. However, generalization to distant regions remains largely unexplored, despite its importance for remote sensing applications. We introduce Matryoshka Implicit Neural Distillation (MIND), which distills embeddings from specialist pretrained geospatial models into a single generalist coordinate embedding with adjustable spatial granularity. MIND uses nested supervision at several embedding dimensions, which define a series of contiguous chunks. In our experiments, early chunks capture coarser geographic variation, while later chunks add more fine-grained details. A downstream predictor can retain only leading chunks or be fitted with our Chunked Penalty to downweight later chunks while keeping the full embedding, without retraining the INR. To measure MIND and compare to existing approaches around the world, we introduce CoordBench, a large-scale INR evaluation suite of 52 datasets and 78 targets that aims to test both local interpolation and prediction in held-out regions at various spatial scales. MIND and its Chunked Penalty variant achieve the highest aggregate regression and classification scores among tested INRs, and the highest scores overall under regional holdout, setting a new state-of-the-art for geographic INRs.
[CV-113] Combinatorial Network-Based Manifold Topological Deep Learning for Image Analysis
链接: https://arxiv.org/abs/2609.25453
作者: Alice Wachira,Xiang Liu,Zhe Su,Yiying Tong,Ge Wang,Guo-Wei Wei
类目: Computer Vision and Pattern Recognition (cs.CV); Biomolecules (q-bio.BM)
备注:
Abstract:Medical image analysis remains fundamentally challenging because of the intricate geometric and topological structures present in medical data. Conventional convolutional neural networks model images as regular Euclidean grids, limiting their ability to preserve geometric relationships and higher-order structural information. Recently, manifold topological deep learning (MTDL) has emerged as a promising paradigm that integrates deep learning with geometric and topological representations. Nevertheless, existing methods have not yet fully exploited discrete manifold structures within combinatorial complex neural networks. To bridge this gap, we introduce CNMTDL, a MTDL framework that integrates Hodge decomposition with a combinatorial attention mechanism. In our approach, medical images are represented as discrete manifolds and decomposed into three Hodge components. Features extracted from these components are concatenated and embedded into a combinatorial complex architecture, enabling enhanced higher-order message passing between 0 -cells and 2 -cells through attention-based blocks. We evaluate CNMTDL on six two-dimensional and three-dimensional datasets from the MedMNIST v2 benchmark, demonstrating its effectiveness for medical image analysis.
[CV-114] Mean Velocity Matching: Rethinking Generative Dynamics in Diffusion Models
链接: https://arxiv.org/abs/2609.25444
作者: Yunhong Zhang,Changjie Cao,Zhihua Zhang,Bingli Liu,Zongjie Cao,Zongyong Cui,Ying Yang
类目: Machine Learning (cs.LG); Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:This work studies prediction parameterization for stochastic generative dynamics in diffusion models. Existing velocity-based generative models provide the simplicity of learning a single transport field, but their standard formulation is deterministic, whereas stochastic extensions generally require additional score information or an intermediate velocity-to-score reconstruction. To retain single-field prediction while directly supporting stochastic reverse dynamics, this paper introduces Mean Velocity Matching (MVM). MVM constructs a Gaussian perturbation process for which the conditional expectation of a restoration-oriented velocity, (x_0-x_t)/t , directly forms the reverse-SDE drift. Consequently, a single learned field is sufficient to parameterize the stochastic reverse process without separately estimating or reconstructing the score. Because direct regression of this velocity becomes unbounded near t=0 , MVM further introduces a \sqrtt -scaled parameterization that preserves the reverse dynamics while yielding a bounded training target. The same learned field also induces a deterministic probability-flow ODE, enabling stochastic and deterministic sampling to be studied within a unified formulation. Experiments with Transformer-based generative models achieve an FID of \MVMImageNetThirtyTwoFID at \MVMImageNetThirtyTwoNFE\ NFE on ImageNet 32\times32 and \MVMImageNetTwoFiftySixFID at \MVMImageNetTwoFiftySixNFE\ NFE on ImageNet 256\times256 . Controlled SDE–ODE comparisons further show that the ODE performs better under very low NFE, whereas the stochastic reverse process achieves lower FID when sufficient function evaluations are available. These results demonstrate that MVM provides a direct single-field parameterization of stochastic reverse dynamics while maintaining competitive generation quality.
[CV-115] Directional Total Variation-Regularized Implicit Neural Representations (DTV-INR) for Continuous Super-Resolution in Degraded Imaging Domains
链接: https://arxiv.org/abs/2609.25429
作者: Mahmoud Saeedi Kelishami
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: 18 pages, 8 figures, 3 tables
Abstract:In this paper, we introduce the Directional Total Variation-Regularized Implicit Neural Representation (DTV-INR), an advanced variational paradigm that synergistically integrates coordinate-driven implicit neural networks with an anisotropic, structure-tensor-informed total variation regularizer tailored for resolution-agnostic image super-resolution. Casting the continuous-to-discrete acquisition process into an ill-posed inverse problem framework, our formulation equips a SIREN-architected coordinate network with a dynamic Riemannian metric tensor field D(x). By leveraging its spectral decomposition, the proposed regularizer preferentially directs diffusion parallel to dominant structural contours while penalizing cross-edge dissipation, successfully circumventing the classical staircasing artifacts inherent to scalar total variation schemes. We rigorously prove the well-posedness of this formulation in H^1(Omega) by establishing the existence, uniqueness, and metric stability of the variational minimizer, and realize this via an alternating projected optimization algorithm that decouples network parameter tuning from adaptive tensor field updates. Comprehensive experiments conducted on clinical brain magnetic resonance imaging (MRI) and biomedical transmission electron microscopy confirm substantial quantitative and qualitative improvements, yielding PSNR enhancements reaching +5.05 dB over baseline unregularized INRs and +1.71-2.85 dB over isotropic TV-INR across continuous (non-integer) upsampling factors, alongside remarkable noise robustness up to sigma_eta = 0.10 and monotonic preconditioned convergence behavior.
[CV-116] Sex Estimation from Footwear Outsole Impressions Using CNN Transfer Learning and Interpretable Image Statistics
链接: https://arxiv.org/abs/2609.25386
作者: Jinyi Niu,Ziyi Song,Weining Shen
类目: Computer Vision and Pattern Recognition (cs.CV); Machine Learning (cs.LG)
备注:
Abstract:Footwear outsole impressions are a common form of forensic pattern evidence, yet quantitative methods for estimating wearer attributes from these images remain relatively underdeveloped. We investigate binary sex estimation from footwear outsole impressions by comparing convolutional neural network (CNN) transfer learning with traditional feature-based classification. Using a publicly available outsole-impression dataset, we adopt a shoe-level training and test partition that keeps replicate scans of the same physical shoe together to reduce data leakage. We evaluate pretrained CNNs through end-to-end fine-tuning, frozen feature extraction followed by support vector machine classification, and hybrid feature fusion incorporating handcrafted, geometric, and metadata-derived descriptors. Fine-tuned CNNs achieve the strongest overall predictive performance and substantially outperform traditional classifiers trained on the manually specified descriptors alone, while frozen-feature approaches offer a less computationally demanding alternative. Exploratory analysis of low-dimensional CNN representations reveals associations with frequency threshold ratio, image contrast, and wavelet-based summaries, providing a connection between learned representations and measurable properties of outsole impressions. These findings suggest that CNN transfer learning captures discriminative information beyond the descriptors considered and offers a promising approach to footwear-based forensic screening. Further validation on independently collected and casework-like impressions is needed before operational use.
[CV-117] PARTE: Plane-Assisted Robust Transformation Estimation for Point Cloud Registration
链接: https://arxiv.org/abs/2609.25375
作者: Abolfazl Babanazari,Carson Cramer,Tyler Summers,Carlos Nieto,Kaveh Fathian
类目: Robotics (cs.RO); Computer Vision and Pattern Recognition (cs.CV)
备注: 16 pages, 14 figures, 11 tables. Code: this https URL
Abstract:Global point-cloud registration remains challenging when limited overlap, repetitive geometry, and sensor noise produce correspondence sets dominated by outliers. Planar regions are particularly difficult for conventional point descriptors and are therefore often suppressed or discarded before matching. We present PARTE (Plane-Assisted Robust Transformation Estimation), a global registration method that instead treats planar structure as complementary registration evidence. PARTE extracts planar patches and represents them using our novel Plane Context Histogram (PCH), a descriptor that encodes the geometry surrounding each patch, while a two-level matching procedure identifies reliable plane correspondences. Candidate point and plane correspondences are combined in a confidence-weighted compatibility graph for joint outlier rejection, followed by rigid transformation estimation. When no usable plane correspondences are available, PARTE naturally reduces to point-only registration. We evaluate PARTE on 8,097 registration pairs across six indoor and outdoor benchmarks spanning dense RGB-D and sparse LiDAR measurements. Evaluations show PARTE achieves the highest overall success rate against 13 standard and state-of-the-art methods while maintaining low runtime. An open-source C++ implementation with Python bindings is provided at this https URL.
[CV-118] MT-ProtBERT: Multi-task Learning ProtBERT for Intrinsically Disordered Proteins Classification with Scarce Data
链接: https://arxiv.org/abs/2609.25334
作者: Jian Sun,Kingshuk Ghosh,Lilianna Houston,Mohammad H. Mahoor
类目: Machine Learning (cs.LG); Computer Vision and Pattern Recognition (cs.CV)
备注: 14 pages, 6 figures, 12 tables
Abstract:Intrinsically disordered proteins (IDPs) differ from folded proteins in that they are dynamic, lack a stable three-dimensional conformation, and have low sequence similarity between similar proteins. The conformational heterogeneity of IDPs - while beneficial for their diverse functions - limits the use of traditional experimental tools to determine their conformation. The experimental difficulty, along with low sequence similarity, results in data scarcity, and makes it difficult to classify/detect IDPs that are similar or dissimilar, a task relevant to understand biology and evolution. We address this challenge using Multi-task ProtBERT (MT-ProtBERT), a multi-task extension of ProtBERT tailored for low-data regimes. MT-ProtBERT integrates Dynamic Window Masking, a Multi-Scale 1D Convolutional classifier (MS-Conv1D), and auxiliary objectives that jointly optimize masked language modeling and biochemistry-informed tasks. We evaluate this framework on two tasks under limited data: (i) phosphorylation site prediction (S/T/Y) in short sequences and small datasets, and (ii) protein compaction prediction on two small datasets (684 and 530 sequences), including sequences comparable in length to typical disordered regions. MT-ProtBERT consistently outperforms PARROT, an RNN-based IDP-specific model, across all tasks. These results demonstrate that combining self-supervised and biochemistry-informed tasks, and multi-scale learning enables robust modeling of unstructured proteins under data scarcity.
[CV-119] MirrorDistill: Illumination-Aware Latent Distillation for Efficient Low-Light Restoration
链接: https://arxiv.org/abs/2609.25331
作者: Farida Mohsen,Tala Zaim,Nurul Izni Rusli,Ali Al-Zawqari,Ali Safa,Samir Brahim Belhaouari
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Low-light image enhancement (LLIE) is an im- portant component of visual sensing systems operating under degraded illumination, including nighttime surveillance, au- tonomous navigation, remote sensing, and inspection in poorly lit industrial environments. Most LLIE methods rely on output- level reconstruction losses that supervise only the final restored image, leaving the intermediate feature recovery process weakly constrained. This paper proposes MirrorDistill, an illumination- aware latent distillation framework that links the low-light and clean domains through feature mirroring. During training, a shared encoder and an exponential-moving-average teacher decoder process the clean reference image to generate clean- domain latent targets. These targets supervise the low-light student at two levels: raw encoder features and standardized multi-scale decoder projections. The alignment is applied layer by layer, while a proposed illumination-aware weighting scheme gives greater emphasis to underexposed regions. The teacher and reference branches are used only during training, so inference requires only the lightweight student encoder-decoder and in- troduces no teacher-side computational cost. Under evaluation on the standard LOL benchmarks, MirrorDistill outperforms the state-of-the-art methods on the real-captured LOL-v2-Real set, while having the lowest compute complexity (GMACs) and while remaining competitive on the LOL-v1 and LOL-v2-Synthetic datasets. Ablation studies further show the contributions of the encoder mirror, decoder mirror, and illumination-aware weighting. Finally, we release our code as open-source for the benefit of future research.
[CV-120] Uncertainty-Aware 3D Residual Wavelet Diffusion for Ultra Low-Field MRI Super-Resolution MICCAI2026
链接: https://arxiv.org/abs/2609.25319
作者: Rui W. Yeow,Millie Beament,Fred Dick,Raha Razin,Martina Bocchetta,David L. Thomas,Henry F. J. Tregidgo,Daniel C. Alexander,James H. Cole
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: 11 pages, 3 figures, 1 table. Accepted at SASHIMI 2026 (MICCAI 2026 workshop). This is the version submitted for peer review
Abstract:Ultra low-field MRI expands global access to neuroimaging but produces scans with low signal-to-noise ratio, reduced contrast, and thick slices. While regression-based super-resolution can recover anatomical detail for segmentation, it returns a single deterministic estimate that gives no indication of regions where the low-field input leaves anatomy underdetermined. Generative diffusion models offer an alternative by sampling the posterior distribution of plausible high-field images, quantifying this anatomical ambiguity. However, applying them to 3D whole-brain MRI is restricted by memory bottlenecks, slow sampling, and scanner domain shifts. We propose a 3D residual wavelet diffusion model that combines three ideas to overcome these hurdles. A lossless wavelet reparameterisation shrinks the spatial grid to fit a whole brain on a single GPU, residual shifting accelerates sampling by starting from the low-field input, and domain randomisation promotes scanner generalisation without paired training data. As the high-field reference is not a voxel-aligned ground truth, we evaluate downstream volumetric agreement. On a healthy cohort (n=19) imaged at 0.064T and 3T, our method matches a leading general-purpose regression approach in volumetric accuracy while additionally generating per-voxel uncertainty maps highlighting underdetermined regions. Furthermore, on a pilot dataset (n=11) of participants with cognitive impairment, disease-relevant atrophy is preserved rather than normalised towards a healthy prior. Our framework brings whole-brain posterior sampling to low-field super-resolution without sacrificing volumetric accuracy.
[CV-121] Beyond the Flat Seafloor: A Closed-Form Two-View Constraint to Aid Sidescan Sonar Reconstruction
链接: https://arxiv.org/abs/2609.25271
作者: Kalin Norman,Joshua G. Mangelson
类目: Robotics (cs.RO); Computer Vision and Pattern Recognition (cs.CV)
备注: Copyright 2026 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media. This work was accepted for publication in the Proceedings of IEEE OCEANS 2026 Monterey
Abstract:Sidescan sonar is a common sensor for both manned and autonomous marine exploration and mapping, yet very few methods build upon or exploit the geometric projection model of the sensor. As sidescan sonar is limited to a 1D range measurement, many approximations are frequently used, including the long-standing assumption of a flat seafloor. Rather than make similar approximations, this paper focuses on a multi-view geometry based approach and formalizes a two-view geometric constraint and proves that a shared feature is constrained to a locus within the intersection of a sphere and a plane. In addition, we characterize what governs the size of the ambiguity locus through Monte Carlo simulation that is grounded in real aperture and mounting geometry for both a surface vessel and an underwater vehicle. We translate additional simulations of relative trajectories for both vehicle platforms into concrete survey-planning guidance. Our results show that locus length is strongly governed by elevation misalignment, and peaks at a moderate oblique crossing angle of approximately 20 degrees, with minimal locus lengths obtained at near parallel and anti-parallel passes.
[CV-122] RULER: Instance-aware Rubric Rewards for SVG Generation
链接: https://arxiv.org/abs/2609.25270
作者: Hangyu Ran,Yuhao Zheng,Yingying Zhang,Kevin Qinghong Lin,Han Peng
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Generating Scalable Vector Graphics (SVG) code from natural-language instructions is an open-ended task without absolute visual ground truth, leaving both evaluation and policy optimization without a faithful signal. Scalar metrics (CLIP, Aesthetic) calibrated on natural images transfer poorly to stylized vector content, and reusing them as RL rewards triggers reward hacking. We address both limitations with rubric-based scoring. We first establish empirically that prompting a vision-language judge with a multi-axis rubric correlates with human judgments far better than scalar metrics, both across samples and within instructions. Building on this finding, we introduce RULER (Instance-aware Rubric Rewards for Reinforcement Learning), which converts each instruction into an instance-aware rubric of six items spanning semantic, visual, and stylistic axes; a judge VLM scores rendered rollouts item-by-item, and the weighted satisfactions form a fine-grained reward optimized via Group Relative Policy Optimization. Because the rubric is derived from text alone, RULER requires neither paired SVG ground truth nor human preference labels. On MMSVG-Illustration and MMSVG-Icon, RULER lifts the rubric score from 0.432/0.395 to 0.693/0.683, surpassing dedicated SVG specialists and matching the substantially larger DeepSeek-V3, with ablations identifying rubric design as the active lever for RL on open-ended SVG generation. The project page is available at this https URL.
[CV-123] ImIR: Image-Instruction Tuning for All-in-One Image Restoration ACCV2026
链接: https://arxiv.org/abs/2609.25267
作者: Süleyman Aslan,Görkay Aydemir,Mısra Yavuz,Yunus Bilge Kurt,Nasrin Rahimi,Ahmet Rasim Emirdağı,Burak Can Biner,M. Akın Yılmaz
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: Accepted to ACCV 2026
Abstract:Degradations vary widely across images, so a practical restoration system has to handle many degradation types with one model. A recent and effective recipe adapts a large pretrained image-editing model to restoration using a small low-rank adapter with a text prompt. We replace that prompt with an instruction derived from the degraded image itself. The image reaches the editor through two paths: its structure comes from the model’s VAE, and its semantic instruction comes from a lightweight token mapper that shifts the degraded image’s vision-language embedding toward the embedding a clean image would produce. Because the instruction is a continuous vector, scaling it yields a family of valid restorations for tasks whose target is not unique, such as low-light enhancement. We adapt one Qwen-Image-Edit model to six tasks with a single adapter trained in about three hours on one GPU. The image instruction outperforms text conditioning under a matched comparison, and it supports task agnostic restoration without a degradation label, which the text variant does not.
[CV-124] Geometric and Semantic Coupling for Interaction Understanding in 3D Scenes
链接: https://arxiv.org/abs/2609.25247
作者: Hanyang Kong,Xingyi Yang
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
备注: Project page: this https URL
Abstract:Interaction understanding in 3D scenes requires a joint description of movable parts, their motion, and the regions through which they can be operated. We present Segment-Snap, which connects these outputs through the physical relationship between parts and handles. Learned predictors identify broad part surfaces and small handles. A geometric decoder uses planar and upright priors to constrain motion, then selects hinge lines using predicted handle locations, without training a motion regressor. Conversely, a joint part-and-handle predictor supplies additional handle candidates, whose motion classes are refined using containing parts. Each information transfer is applied once, without iterative feedback. On Articulate3D validation, handle guidance raises motion-gated AP from 13.74% to 40.98% at fixed masks and axes. Additional handle candidates raise handle AP from 24.63% to 29.65%; part-based class correction adds 0.98 points, and full context reaches 30.99%. Repeated training, learned-decoder controls and paired visualizations establish the benefits and limitations of combining geometric and semantic evidence for interaction understanding.
[CV-125] Youve Seen Enough: Quality-Constrained Image Coding for Machines
链接: https://arxiv.org/abs/2609.25108
作者: Khoa Pham-Dinh,Sanaz Nami,Hamed Rezazadegan Tavakoli,Moncef Gabbouj,Farhad Pakdaman
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
备注:
Abstract:Visual data is increasingly consumed by machine-vision systems rather than by human observers. Image Coding for Machines (ICM) compresses images assuming the main observer is a computer vision application and that the human observer needs to inspect or validate the decisions. Inspired by just-noticeable distortion, which sets the quality to the just-acceptable level for human observers, we aim to cap the human-observed quality at a desired level, with the goal of using the remaining coding capacity to improve the machine performance. We recast joint compression-segmentation training as a constrained optimization problem in which the codec must meet a predefined acceptable target visual quality while a task term consumes the remaining coding capacity. We solve this by designing a penalty function to guide the quality to the desired target. We propose two penalty functions, an absolute function and a bilinear function, the latter applying a steeper slope once the target visual quality is exceeded. Experimental results show that, under the quality constraint, the proposed method achieves a BD-rate of -22.82% over an unconstrained joint rate–distortion–task optimization and -29.81% over a simple rate–distortion baseline, showcasing bitrate reduction with the same task performance. This is achieved while the codec also meets the target visual quality with a reasonable error and without adding any complexity overhead.
[CV-126] SPARC: SuperPixel-Aware Region Contrastive Learning for Self-Supervised Dense Prediction ICASSP2027
链接: https://arxiv.org/abs/2609.25067
作者: David Szczecina,Yuanpei Xiang,Jitao Hu,David Clausi,Yuhao Chen,Jason Deglint,Paul Fieguth
类目: Computer Vision and Pattern Recognition (cs.CV); Machine Learning (cs.LG)
备注: 5 pages, 2 figures. Submitted to the IEEE ICASSP 2027 Conference
Abstract:Self-supervised learning (SSL) has become an effective approach for learning visual representations without manual annotations. Among SSL approaches, contrastive learning has been widely used for visual representation learning. However, existing contrastive SSL methods have focused primarily on image-level or pixel-level representation learning, while region-level representation learning remains less explored. We propose SPARC, a region-level contrastive learning framework that leverages superpixels to establish explicit correspondence between augmented image views. SPARC introduces a region contrastive branch that performs superpixel-based feature pooling and optimizes a region-level contrastive objective jointly with a global image-level objective. Under identical settings, SPARC consistently outperforms previous methods such as MoCo-v2 and DenseCL, achieving improvements of up to +9.79 mIoU for semantic segmentation and +4.88 AP for object detection. Ablation studies further demonstrate that region-level objectives produce the strongest performance. Thus, region-level contrastive learning is an effective approach for improving self-supervised visual pretraining for dense prediction tasks. Code repository can be accessed at this https URL.
[CV-127] NPLSD: Accelerating Line-Segment Detection on NPU Microcontrollers
链接: https://arxiv.org/abs/2609.25022
作者: Parsa Hassani Shariat Panahi,Amir Hossein Jalilvand,M. Hassan Najafi
类目: Hardware Architecture (cs.AR); Computer Vision and Pattern Recognition (cs.CV)
备注: 7 pages, 5 figs
Abstract:Line-segment detection is fundamental to robotics, autonomous navigation, and industrial inspection. While transformer-based detectors achieve the highest accuracy, their deployment on microcontrollers remains impractical due to resource constraints. The STM32N6, with its Neural-ART NPU, promises to enable deep vision at the extreme edge. However, existing detectors rely on attention, grid-sampling, and normalization, operators that are unsupported by the convolution-oriented NPU. This architectural mismatch is characterized operator by operator: attention, grid-sampling, and normalization lack accelerator primitives, and the decoder’s self-attention alone materializes a 39 MB score tensor that exceeds on-chip memory. To address this limitation, NPLSD is introduced as a pair of NPU-compatible line-segment detectors built from one design methodology. NPLSD-H retains the convolutional HGNetv2 backbone of LINEA and replaces the transformer head with a fully-convolutional feature pyramid and an F-Clip dense head. NPLSD-M adapts the M-LSD-tiny trunk to the supported operator set. Warm-started from ImageNet and trained on ShanghaiTech Wireframe, the 2.63M-parameter NPLSD-H reaches sAP^10=37.9 (35.9 int8); the 0.62M-parameter NPLSD-M reaches 41.9 (41.1 int8). A controlled ablation isolates the trunk as the only variable, and initialization alone accounts for 4.6 points.
[CV-128] Deepfakes and Synthetic Media: Generation Detection and Governance
链接: https://arxiv.org/abs/2609.25017
作者: Alexandros Gazis,Efstathios Karypidis,Kleanthi Santamouri,Theodoros Vavouras,Nikos E. Mastorakis,Stylianos Pappas
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: 30 pages, 2 figures, 3 tables, 120 references
Abstract:Deepfakes, synthetic audiovisual content produced by deep generative models, have escalated into a critical threat across civilian and military domains, enabling identity fraud, disinformation campaigns, and evidence fabrication. In high-stakes environments, ranging from journalism and finance to healthcare and legal contexts, the consequences extend to severe misinformation, market manipulation, identity fraud, and the erosion of institutional trust. This entry explores how modern visual intelligence and computer vision techniques are used to detect deepfakes. It outlines key deepfake generation models, such as GANs, autoencoders, neural rendering, and diffusion systems, while also explaining how adversarial methods enhance realism and challenge existing detectors. The overview highlights visual artifacts, digital patterns, and physiological cues commonly leveraged in detection and reviews major CNN, transformer, and frequency-based approaches. It also summarizes evaluation practices and the difficulty of achieving strong generalization. Finally, it identifies emerging directions, including modern intelligence techniques for civilian and military content verification. This survey covers generation architectures (GANs, latent diffusion, neural rendering, video synthesis), the spatial, temporal, frequency-domain, and physiological artifacts they produce, and the detector families that exploit them. We examine evaluation benchmarks and protocols, highlighting cross-generator generalization as the field’s central open challenge. Beyond detection, we discuss cryptographic provenance standards, watermarking, and regulatory frameworks (EU AI Act, DSA, GDPR). We conclude that effective deepfake governance requires defense-in-depth integrating forensic detection, verifiable provenance, and institutional accountability.
[CV-129] IR: Unlocking Instance-Specific State Evolution via Test-Time Training for Image Restoration
链接: https://arxiv.org/abs/2609.26151
作者: Kaihang Zheng,Jun Li,Hang Guo,Hongyu Chi,Zimo Liu,Tao Dai,Jinpeng Wang,Yaowei Wang
类目: Image and Video Processing (eess.IV); Artificial Intelligence (cs.AI); Computer Vision and Pattern Recognition (cs.CV); Multimedia (cs.MM)
备注: TL;DR: TTTIR improves image restoration by framing it as an instance-specific state evolution process. Powered by Test-Time Training (TTT), it dynamically adapts operators to handle real-world degradations, outperforming state-of-the-art models with scalable efficiency. 11 pages, 6 figures, 6 tables
Abstract:Image restoration is inherently challenging due to the diverse and highly input-dependent nature of real-world degradations. While recent architectures like Transformers and state-space models have advanced the field, they predominantly rely on static, globally shared parameters, which struggle to fully accommodate instance-specific degradation patterns. Test-Time Training (TTT) offers a promising paradigm for generating data-dependent operators, yet its standard self-supervised inner loop lacks the explicit guidance required to transition degraded features toward clean structures. To address this, we propose TTTIR, a novel framework that reformulates image restoration as an instance-specific state evolution process. Specifically, we design Progressive State Sequence Generation (PSSG) to construct complementary spatial-frequency target states (defining what to recover), and State Transition Evolution (STE) to adapt lightweight transition operators via a restoration-oriented TTT inner loop (determining how the features should evolve). Extensive experiments demonstrate that TTTIR consistently outperforms state-of-the-art models across multiple image restoration benchmarks, achieving dynamic instance-specific recovery with favorable computational scalability. The code is available at this https URL.
[CV-130] Robust Estimator-Agnostic Dynamic 3DGS Compression ICASSP2027
链接: https://arxiv.org/abs/2609.25633
作者: Chenjunjie Wang,Zixi Huang,Yao Wang,Jona Ballé
类目: Image and Video Processing (eess.IV); Computer Vision and Pattern Recognition (cs.CV)
备注: Submitted to IEEE ICASSP 2027. This version adds an appendix; 16 pages, 15 figures
Abstract:Dynamic 3D Gaussian splats (3DGS) model time-varying scenes using a separate Gaussian set per frame. While neighboring video frames are highly correlated due to smooth motion, Gaussian representations retain this correlation to varying degrees, depending on whether the estimator tracks them across time. Some 3DGS compression methods integrate the estimation to exploit temporal redundancy; here, we focus on robust compression regardless of the estimator. We concatenate groups of frames into one Gaussian set, augment each Gaussian with a frame index, and pass it to a static (i.e., non-temporal) 3DGS codec, converting temporal redundancy into spatial redundancy. Concatenated sets are spatially partitioned to limit memory. Our technique requires neither a motion model nor knowledge of the training method. Averaged over six N3DV sequences, all six static codecs achieve gains on tracked sets (-42.0% to -71.8% BD-rate) over per-frame coding. On untracked sets, all codecs except HGSC, which appears incompatible with our technique, remain competitive with per-frame coding (-3.5% to +5.0%). We further replace D-FCGS’s I-frame coding with our technique while retaining its P-frame coding, yielding an overall BD-rate of -46.2%. We propose to visualize “trackedness” using an inter-frame similarity metric. The project is available at this https URL.
[CV-131] Calibration Count Reuse: Validity Does Not Determine Efficiency
链接: https://arxiv.org/abs/2609.25138
作者: Rudra Chopra
类目: Methodology (stat.ME); Computer Vision and Pattern Recognition (cs.CV); Machine Learning (cs.LG)
备注: 33 pages, including appendices
Abstract:Calibration count reuse raises separate validity and efficiency questions. We give a validity criterion for general count-dependent nonconformity scores: transferring one count from another class to the scored class must not improve its conformity. A leave-self-out full conformal reference proves the criterion without requiring normalization or preservation of same-class score order. For a common separable transformation, universal exchangeable validity is equivalent to being nondecreasing in the count, provided K\alpha \geq 1 ; normalized multiplicative weights obey the complementary nonincreasing condition. Additive penalties are covered under the stated information restrictions. Efficiency has no parallel ordering: two iid constructions make the same valid rule improve or worsen expected size at unchanged coverage. An expanded 55-rule study finds no resolved advantage from selected live-count rules over uniform weights. Image studies identify undercoverage under iid resampling, including at numerical convergence. Separately, execution of the released Conf-OT pipeline on its DTD and Aircraft benchmark subsets produces near-nominal median coverage under fixed stratified counts. The native results are reported separately from the iid analyses, without treating a benchmark observation as a universal guarantee. The findings separate validity, classifier confidence, numerical convergence, and population-specific efficiency.
[CV-132] WILSON - a pathology foundation model framework for patient-level analysis and diagnostic text generation
链接: https://arxiv.org/abs/2609.25123
作者: Saghir Alfasly,Wataru Uegami,Sobhan Hemati,Wenchao Han,Xiaojia Tang,Kevin Thompson,Daniel Stone,Ghazal Alabtah,Saba Yasir,Michael R. Lucas,Eric W. Klee,Cheryl L. Willman,Judy C. Boughey,Matthew P. Goetz,Krishna R. Kalari,H.R. Tizhoosh
类目: Quantitative Methods (q-bio.QM); Artificial Intelligence (cs.AI); Computer Vision and Pattern Recognition (cs.CV); Machine Learning (cs.LG); Image and Video Processing (eess.IV)
备注: 56 pages, 6 main figures, with 11 additional figures and 28 tables in the appendices
Abstract:Pathologists integrate morphology across magnifications and across the slides of a patient case, whereas pathology foundation models encode thousands of tiles from single slides and aggregate their features. Here we present WILSON, a vision–language foundation model that represents whole-slide images and multi-slide cases as single multi-magnification composite images, trained on approximately 189k slides from Mayo Clinic spanning 42 organs and 829 diagnostic entities using pathology reports as supervision. Without task-specific training, WILSON exceeded a dedicated case-level model on all internal cohorts (macro-F1 0.52 versus 0.38) and matched slide-level models up to 9.4 times larger at 272- to 2,155-fold lower compute. End-to-end fine-tuning on 508 triple-negative breast cancer cases improved histologic subtyping and stromal tumor-infiltrating lymphocyte grading by 0.16 and 0.11 macro-F1. WILSON retrieved matching diagnostic text at 75.6% recall@1 (PRISM, 58.1%) and generated captions closer to report-derived references than PRISM and PRISM2 on the internal cohort and on most external comparisons. Composite images thus offer a compact, clinically aligned computational unit for pathology.
[CV-133] BananaVLM: A Domain-Adapted Vision Language Model for Banana Crop Disease Diagnosis
链接: https://arxiv.org/abs/2609.25040
作者: Sangam Kumar Jena,Pandarasamy Arjunan
类目: Quantitative Methods (q-bio.QM); Computer Vision and Pattern Recognition (cs.CV)
备注: 11 pages
Abstract:Banana crop diseases threaten food security across the world, yet field diagnosis remains difficult because of limited expert access and visual similarity among disease classes. We present BananaVLM, a domain-adapted vision-language model for banana disease diagnosis built on LLaVA-v1.5-7B and fine-tuned with Low-Rank Adaptation (LoRA). To address the lack of multimodal agricultural instruction data, we introduce BananaInstruct, an automated pipeline that converts raw disease images into \approx 80,000 question–answer pairs without manual annotation, using LLaVA-1.5-13B for symptom descriptions and Mistral-7B for agricultural QA generation and label grounding. Evaluated against 14 open-source and 5 closed-source VLMs on a nine-class benchmark, BananaVLM achieves 92.21% in-domain and 83.28% out-of-domain (OOD) classification accuracy, outperforming the best closed-source baseline, Gemini~2.5~Pro, by 49.8 and 63.3 percentage points, respectively. Binary healthy/diseased identification reaches 98.38% OOD accuracy with 1.00 recall. A controlled LoRA–DoRA comparison shows that DoRA performance degrades across epochs, while LoRA remains more effective for fine-grained classification. Qualitative evaluation using nine LLM judges yields G-Eval win rates of 0.81–0.96, while five domain experts prefer BananaVLM in 98.48% of 990 blind pairwise comparisons. These results show that lightweight domain adaptation with automated instruction tuning is an effective and scalable approach for specialized agricultural AI. Code, datasets, and model weights are available at this https URL
人工智能
[AI-0] CliffCompaction: Cost-Efficient Compaction for Long-Horizon Coding Agents
链接: https://arxiv.org/abs/2609.26779
作者: Trang Nguyen,Eulrang Cho,Bingqing Chen,Tim Dettmers
类目: Artificial Intelligence (cs.AI); Machine Learning (cs.LG); Software Engineering (cs.SE)
备注:
Abstract:Agents often work on complex problems that require millions of tokens of context, which necessitates compacting across sessions due to limited context windows. We develop CliffCompaction, an autocompaction technique that reduces cost by up to 50% under a bounded context while maintaining or improving performance on Terminal-Bench and achieving new levels of efficiency for test-time scaling and state-of-the-art results on KernelBench. The per-rollout savings of CliffCompaction make the performance–cost trade-off of test-time scaling more efficient, adding over 10 percentage points on Terminal-Bench for less than the cost of two full-context runs. Under parallel test-time scaling, CliffCompaction lets Kimi K2.6 match Opus 4.7, and exceed Opus 4.6 and GPT-5.3 Codex at lower cost. The key to CliffCompaction’s effectiveness is that it keeps compacted information faithful by only truncating or dropping content, never rephrasing or rewriting it. We never compact a compaction—each pass operates only on original content, and prior compacted output is discarded, preventing context drift from accumulating. These properties sustain continual learning over sessions exceeding a million tokens: on KernelBench, CliffCompaction reaches CUDA kernel speedups of 2.23\times after 200 steps and 3.58\times after 400 steps, surpassing specialized search algorithms and trained agents despite being a general-purpose compaction technique. We open-source a scaffold-agnostic API-proxy implementation of CliffCompaction usable with Claude Code, Codex and other harnesses.
[AI-1] SWE-Serve: Benchmarking Agent ic Engineering For Production Inference Serving
链接: https://arxiv.org/abs/2609.26777
作者: Jennifer Williams,Dave Farris,Jeff Farris,Jiantao Jiao
类目: Artificial Intelligence (cs.AI); Software Engineering (cs.SE)
备注:
Abstract:We introduce SWE-Serve, a benchmark for evaluating agents on production inference engineering tasks. Implementing an inference feature can require coordinating multiple changes across the serving stack, including model support, runtime execution, and public APIs. Existing benchmarks provide limited coverage of production inference engineering: repository-level software engineering benchmarks do not target inference, while general terminal-agent benchmarks include only a few inference tasks. Dedicated inference benchmarks, meanwhile, focus primarily on isolated kernel generation or performance optimization rather than repository-scale production feature implementation. SWE-Serve provides 53 repository-grounded tasks derived from recent production changes to SGLang, spanning six inference engineering families. Each task executes on either CPU or a single GPU (H100) and is evaluated with hidden functional and regression tests, including, where applicable, end-to-end (E2E) serving tests and calibrated performance gates. Executable no-op and oracle controls, adversarial verifier review, and closed-book execution support task validity and evaluation integrity. Across 11 models and 31 model-effort configurations, the best-performing configuration achieves 75% mean pass@1. SWE-Serve exposes a substantial gap between completing tasks locally and achieving production correctness. On 19 tasks with end-to-end coverage, model-serving E2E tests reject roughly one-third of patches that pass every other test (45.9% under the verifier versus 69.4% with E2E tests excluded from scoring), with pass rate increasing for each model’s best-performing configuration. By making the production correctness gap directly measurable, SWE-Serve enables the field to track whether future agents move beyond completing tasks locally to achieving production correctness.
[AI-2] A2M: Trace-Optimized Agent Hijacking in the MCP Ecosystem AACL
链接: https://arxiv.org/abs/2609.26761
作者: Laizhen Li,Xuan Wang,Peicheng Zhao,Juanjuan Zhao,Kejiang Ye,Cheng-zhong Xu,Xitong Gao
类目: Cryptography and Security (cs.CR); Artificial Intelligence (cs.AI)
备注: Accepted by AACL-IJCNLP 2026
Abstract:Agents using the Model Context Protocol (MCP) rely on semantic matching to select tools from third-party servers, exposing a semantic supply-chain risk through attacker-controlled metadata and outputs. We introduce A2M (Attraction-to-Manipulation), a two-stage black-box framework for hijacking MCP agents. The Attraction phase optimizes tool metadata to increase invocation probability; the Manipulation phase uses execution traces to refine adversarial tool returns that steer agents toward attacker-desired outcomes. On LiveMCPBench, direct attacks optimized and evaluated on GLM-4.6 achieve a macro-average malicious tool invocation rate of 93.6% across four scenarios, increase weighted token costs to 32.4 \times the benign baseline under Cognitive Denial of Service, and attain a mean attack success rate of 74.4% across Information Exfiltration, Environment Integrity Compromise, and Reasoning Derailment. Transfer to four other models without re-optimization yields corresponding macro-averages of 63.6%, 2.7 \times , and 24.5%. These findings motivate stronger tool vetting and runtime isolation in MCP ecosystems. Code is publicly available at this https URL.
[AI-3] Grow the Harness Not the Context: From Strategy-Free Scaffolds to Reusable Specialist Agents
链接: https://arxiv.org/abs/2609.26760
作者: Laizhen Li,Jiarui Li,Juanjuan Zhao,Kejiang Ye,Ye Li,Cheng-zhong Xu,Xitong Gao
类目: Artificial Intelligence (cs.AI); Software Engineering (cs.SE)
备注: 16 pages, 6 figures
Abstract:Large language model (LLM) agents often handle streams of related tasks, yet standard harnesses repeatedly ask the model to reconstruct the same control decisions inside each task’s context. We study whether task feedback can instead turn recurring control into reusable executable code, while reserving LLM calls for task-specific semantic reasoning. We introduce Growing Harness, a failure-guided training paradigm that learns the agent harness itself from a strategy-free scaffold that exposes fixed model and tool interfaces but encodes no task-solving controller. Function-level execution traces localize each failure to a bounded code surface, an optimizer repairs a window of failures jointly, and a success-first held-out gate rolls back repair sequences that harm prior capability. Accepted edits accumulate in one shared harness, allowing its control structure to emerge from task feedback. Across BrowseComp-Plus and WebArena-Verified with three deployment models from 4B to 120B parameters, Growing Harness achieves the highest mean success in five of six benchmark-model settings and trails the best mean by 0.7 pp. in the sixth. Relative to a Tool-Calling agent, it reduces LLM calls by 76.0-91.8% and deployed-agent inference cost by 74.4-98.6%. On WebArena-Verified, its success remains 44.7-45.3% across model scales, whereas Tool-Calling falls to 6.7% with the 4B model. Ablations show that trace-local edits, joint repair, and gate-based rollback each improve final success. These results show that persistent program growth can move recurring control out of model context and into low-cost code, yielding reusable specialist agents that remain effective with smaller deployment models.
[AI-4] ype-Safe Is Not Error-Free: A Constrained Decision Head Follows the Option Name Not the Rubric Bound to It
链接: https://arxiv.org/abs/2609.26758
作者: Yu Sun,Junhao Xu
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Typed decision models are built for settings where model outputs are consumed directly by software. Instead of generating free-form text, they return a decision over a predefined set of options. By construction, every output conforms to the required schema. Yet this guarantee does not tell us whether the model interprets the options as intended. We study Jev and two Jev-like models with open weights by changing how option names are assigned to rubrics. Each option consists of an option name and a textual rubric that defines what the option means. We change only which option name is assigned to each rubric; the question, state, rubric wording, and set of option names remain exactly the same. On 1200 workflow decisions with task-specific rubrics, renaming the two options from 0/1 to no/yes changes 70.4 more answers per hundred (95% CI: [67.6, 73.1]) and shifts AUC from .94 to .23, revealing a systematic reversal in the decision ranking rather than simple uncertainty. The same operation has little effect with neutral option names. This pattern holds across all 4 predicates, where the effect is at least 7.4x larger than under the neutral control, and becomes stronger as the number of options increases. The effect also depends on the read-out geometry: a second model family that mean-pools over the full option span flips 4.1x less often. The hosted model exhibits the same behavior: the swap changes AUC from .8146 to .5806 and produces 24x as many answer flips as its test-retest floor. In contrast, replacing the option names with random character strings returns all model families to the neutral-control regime without reducing accuracy. The failure therefore depends on the semantic polarity of the option names rather than on the renaming operation itself. Across all conditions, the type-error rate remains 0%, even when decision accuracy degrades substantially.
[AI-5] Metrics Failure in LLM -Based Code Vulnerability Repair: An Empirical Study and a Change-Aware Screen
链接: https://arxiv.org/abs/2609.26749
作者: Om Nepal,Sushant Aryal,Oluseyi Olukola,Nick Rahimi
类目: oftware Engineering (cs.SE); Artificial Intelligence (cs.AI); Cryptography and Security (cs.CR)
备注: 23 pages, 4 figures, 11 tables. Code and data: this https URL
Abstract:Large language models (LLMs) are increasingly applied to the automated repair of C/C++ security vulnerabilities, and compile rate is a commonly reported proxy for progress: whether the generated patch compiles. We argue that compile rate is a scientifically unreliable metric for single-function vulnerability repair, and we support this with five controlled experiments over 203 vulnerable functions from Big-Vul, three open-source code LLMs (350M to 6.7B parameters), and three prompting strategies. Compile rate (i) barely responds to an intervention that substantially improves the generated code; (ii) is dominated by evaluation-harness and dataset artifacts rather than model quality, with about 64% of compile failures not attributable to the model, a share that is nearly invariant across models; (iii) shifts by 1.8 to 2.7 times on identical patches under a single compiler-standard flag, with zero regressions; (iv) ranks the three models in the opposite order to reference-similarity metrics; and (v) rewards non-repairs when used as an optimization target, since a compiler-feedback loop raises compile rate while similarity to the human fix falls, with manual inspection finding deletion- and placeholder-style non-repairs among the newly compiling outputs. The natural fallback, whole-function CodeBLEU, also fails: an unchanged copy of the vulnerable input outscores every model. We also examine diff_F1, a change-aware screen that scores only the edited region. It gives exactly zero credit to a no-op and near-zero credit to some, though not all, of the deletion-based gaming patches we observed, while still crediting genuine partial edits, so it may serve as a cheap screen before deeper, execution-based analysis. It is not a repair-quality metric, and we report where it falls short. Our findings argue for change-aware, execution-grounded evaluation of LLM-based vulnerability repair.
[AI-6] he Sirens Song: When Proximal Background Context Overshadows Distant Evidence
链接: https://arxiv.org/abs/2609.26718
作者: Xiaoyu Yang,Jie Lu,Wei Duan,En Yu
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
备注: 18 pages
Abstract:Long-context LLMs focus on retrieving distant evidence from extensive context, yet existing work has largely focused on overcoming distance alone. In this work, we identify the Proximity Trap, insufficient attention to distant evidence often arises less from distance itself than from cumulative competition with abundant, task-irrelevant proximal background. To address the Proximity Trap, we introduce LYRA (Long-context heavY-tailed Relevance Alignment), a t-distributed directional matching mechanism that reshapes the context retrieval distribution, directing more attention mass toward task-relevant evidence, while preserving the relative positional information encoded. Extensive experiments on LongBench-v2, RULER, and LongBench demonstrate consistent improvements across context lengths and task categories. We further introduce ProxBench, a multi-level fine-grained benchmark for evaluating distant evidence utilization under increasing proximal background interference. Project page: this https URL
[AI-7] raceVIC: Causal Reasoning over Code Evolution for Identifying Vulnerability-Inducing Commits
链接: https://arxiv.org/abs/2609.26711
作者: Fnu Tanish,Samiha Shimmi,Samikshya Chapagain,Hamed Okhravi,Mona Rahimi,Lei Zhang
类目: oftware Engineering (cs.SE); Artificial Intelligence (cs.AI)
备注:
Abstract:Software vulnerabilities are often discovered long after they are introduced, making it difficult to identify the vulnerability-inducing commit (VIC) responsible for introducing the underlying vulnerable condition. Existing VIC identification techniques largely rely on git blame to trace vulnerable code through revision history and use positional heuristics, such as selecting its earliest or most recent modification. However, the true VIC may occur anywhere within this history, and vulnerable behavior may depend on code that evolves across multiple revisions. We therefore argue that VIC identification requires reasoning about how vulnerability-relevant code evolves, rather than simply where a candidate commit appears in the revision history. We present TraceVIC, a temporal graph-based approach for identifying and ranking VICs by reasoning over code evolution. TraceVIC first localizes likely root-cause lines and traces their histories across revisions, constructing graph representations that capture program structure within each revision and the evolution of vulnerability-relevant code across the history. It reasons over the resulting revision history, using temporal edges to preserve correspondences between program elements across consecutive revisions, and directly ranks candidate commits according to their contribution to the vulnerable condition. Ablation results show that modeling the full revision history improves F2 from 0.637 to 0.814. TraceVIC improves F2 by up to 28.7% over state-of-the-art methods and identifies a valid VIC for 78 of 79 vulnerabilities across four unseen C/C++ projects. Subjects: Software Engineering (cs.SE); Artificial Intelligence (cs.AI) Cite as: arXiv:2609.26711 [cs.SE] (or arXiv:2609.26711v1 [cs.SE] for this version) https://doi.org/10.48550/arXiv.2609.26711 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
[AI-8] rain Where the Quantized Model Goes: On-Policy Distillation for Low-Bit Reasoning
链接: https://arxiv.org/abs/2609.26708
作者: Yuanteng Chen,Zhilei Liu,Peisong Wang,Yuantian Shao,Chuangyi Li,Weining Wang,Shuang Qiu,Gang Li,Jing Liu,Jian Cheng
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
备注: 18 pages, 6 figures
Abstract:Quantization-aware distillation (QAD) restores much of the short-form question-answering performance lost to sub-3-bit quantization, yet leaves mathematical and code reasoning substantially impaired. Long generations often degenerate into repetitive loops, exhausting the decoding budget without completing a solution. We trace this gap to quantization-amplified exposure bias: QAD trains on fixed corpus prefixes, while quantization-induced deviations compound along the model’s own autoregressive trajectories. To address this mismatch, we introduce an on-policy distillation (OPD) stage that places teacher supervision where the quantized model actually goes. Starting from a QAD checkpoint, the student generates through the quantized forward path used at deployment and receives feedback from a frozen full-precision teacher on its own prefixes, combining dense token-level guidance with task-verifier rewards. Across four models at 2.79 and 1.88 effective bits, OPD raises average BF16 performance retention from 35% to 70% on MATH-500 and from 66% to 91% on HumanEval while preserving short-form performance, with reasoning gains substantially exceeding those of continued teacher-forced QAD in matched-budget comparisons. By coupling QAD’s stable low-bit initialization with OPD’s on-policy reasoning recovery, our framework provides a comprehensive sub-3-bit solution that preserves broad capabilities while restoring long-form reasoning.
[AI-9] From Alignment to Access Control: A Framework for GenAI Policy Enforcement
链接: https://arxiv.org/abs/2609.26682
作者: Nathalie Baracaldo
类目: Cryptography and Security (cs.CR); Artificial Intelligence (cs.AI)
备注:
Abstract:Generative AI (GenAI) applications have flourished enabling users to chat with large language models, and to create agents to act on their behalf for a variety of tasks. The pace of development of capabilities in this field is incredibly fast with security and safety taking a back seat. Unfortunately, the slower pace at which security and safety mechanisms have evolved has led to real incidents. Policy enables the definition of desirable behavior of applications, and for that reason, it is a cornerstone of making systems secure and compliant. Policy however means different things to different practitioners creating confusion and siloed solutions that are not adequate for compliance. This paper takes a tour of the good, the bad and the ugly when it comes to policy enforcement in GenAI applications. We propose a methodology to systematically analyze and dissect existing approaches to define and enforce policy found in the wild. Based on this principled analysis, we provide recommendations and call for action for the community to address. This paper is a companion extension of USENIX Security 2026 Enigma talk titled “From Alignment to Access Control: A Unified View of GenAI Policy Enforcement” by the author Nathalie Baracaldo. Subjects: Cryptography and Security (cs.CR); Artificial Intelligence (cs.AI) Cite as: arXiv:2609.26682 [cs.CR] (or arXiv:2609.26682v1 [cs.CR] for this version) https://doi.org/10.48550/arXiv.2609.26682 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
[AI-10] A Spectral Theory of Grokking: Weight Decay induces Feature Learning
链接: https://arxiv.org/abs/2609.26679
作者: Lenz Pracher,Pascal de Jong,Oskar Lieshaus,Alan Jeffares,Steffen Rulands
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
备注:
Abstract:In grokking an early fit to the training data separates from a much later improvement in generalization. During this delay, training can move from a fixed neural tangent kernel (NTK) regime to one in which task-relevant kernel eigendirections continue to evolve. We provide a quantitative theory for how this transition from lazy to rich learning can produce delayed generalization. For homogeneous networks trained with squared loss and L_2 weight decay, we show that a finite residual remains after memorization, with larger residual fractions in target components associated with smaller NTK eigenvalues. These residuals feed back into the dynamics of the NTK itself, and projecting the resulting dynamics onto task-relevant spectral directions yields a reduced system in which residual-driven kernel growth competes with weight decay. This system predicts that the grokking timescale is controlled by the product of learning rate and weight decay, that feature learning slows logarithmically near a critical decay above which task-aligned NTK structure can no longer support generalization, and that stronger decay can prevent fitting altogether. We test these predictions in modular addition. In a homogeneous MLP, task-aligned Fourier structure continues to emerge in the NTK after training accuracy has saturated, and an 84 \times 90-grid of trained networks across varying learning rate and weight decay recovers the predicted phase geometry and inverse-product scaling of the generalization time with learning rate and weight decay. A one-block Transformer shows similar macroscopic phase structure in a 42 \times 45-grid, as well as the same transition-time scaling despite violating exact homogeneity. Together, these results provide a mechanistic derivation connecting post-fit feature learning to both the onset of generalization and its phase structure in the learning rate and weight decay plane.
[AI-11] he Delegation Blind Spot: Auditing Product Decisions from Agent Choices ICIP
链接: https://arxiv.org/abs/2609.26642
作者: Shivam Gupta
类目: Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
备注: 15 pages, 5 figures. Computational technical report with proofs and synthetic-task experiments; no human participants. Code: this https URL
Abstract:Successful agent execution need not identify which future product improvement its user would value. We present a decision-specific audit that maps a declared observation channel and product-value contrast to compatible intervals and witness populations. Its foundations are established identification and decision theory; the contribution is an executable measurement workflow and a controlled study of its limits. A frozen experiment makes 4,800 requests to two pinned model snapshots on shared synthetic tasks. All 36 conservative primary intervals remain unresolved despite different execution accuracy. An exploratory 2,400-call follow-up records supplied preferences and resolves three of nine comparisons per model. A deterministic extractor resolves seven of nine without model calls or calibration observations, exposing unnecessary uncertainty introduced by model-generated reports. A further 14,400 controlled multinomial simulations distinguish structural ambiguity from weak identification and finite calibration precision. We propose a source-labeled decision receipt and provide an offline viewer for inspecting the audit. These results motivate preserving decision-relevant structured input and diagnosing why a decision is unresolved before collecting more telemetry. The study contains no human participants or real customer outcomes. Full proofs, raw model provenance, controlled experiments, and reproducible analyses accompany the report.
[AI-12] Greedy Decoding Is Not Precision-Invariant: Cross-Precision Output Divergence in LLM Inference
链接: https://arxiv.org/abs/2609.26621
作者: Gaoyuan Du,Anam Nawaz Khan,Rex Zhou,Xiaoyang Liu,Deepayan Chakrabarti,Fnu Suya,Xueping Li
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
备注: Accepted by Transactions on Machine Learning Research (TMLR), 2026
Abstract:Greedy decoding from large language models is commonly treated as deterministic. We show it is not precision-invariant: the same model, prompt, and decoding algorithm produce different outputs in BF16 versus FP16 on identical hardware. Across our evaluations of six models (1.1B-7B parameters, four families; divergence additionally characterised at 12B) and three benchmarks, 49-100% of prompts diverge; a single token flip often cascades into trajectory-level divergence. We develop an empirical error-propagation analysis and find that 22 layers of accumulated body error do not distinguish flipping from non-flipping steps; the outcome depends primarily on the top-two logit margin at the LM head relative to the directional perturbation between the top-two candidates. The analysis makes five testable predictions about intervention outcomes, including that applying more FP32 compute (broader scope) makes agreement worse. The experiments match all five predictions. The best-performing low-overhead intervention we evaluate, selective FP32 LM head recomputation, triggered only when the margin falls below a threshold, delivers +22-36 pp exact agreement on A10G (+12-21 pp on L4 and A100) at less than 4% latency overhead in low-batch (batch size =4) single-stream inference. We map the applicability boundary across six models and four batch sizes, and hypothesise that training-time precision stability is a determining factor. The method is a partial mitigation rather than a universal determinism guarantee: its benefit vanishes when body-originated error dominates, including at batch size =8 and under end-to-end FP8 in our tests.
[AI-13] owards Hierarchical GNNs for multi-grid power flow: generalization across operating scenarios
链接: https://arxiv.org/abs/2609.26603
作者: Carmine Delle Femine,Leire Garin Atxaga,Asier Diaz-Iglesias,Juan Pablo Maroto Herrera,Ane Miren Florez-Tapia,Marco Quartulli. Izaro Goienetxea Urziku
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
备注:
Abstract:Hierarchical latent communication improves the generalization of a multi-grid power-flow model to new operating scenarios. The module exchanges information through two reduced graphs within a GENCO-based corrective network. We compare Kron-derived transports, a same-anchor Quotient construction and a flat backbone in preliminary trainings of 200 epochs on three grid topologies, with three initialization seeds per model. Evaluation uses 200 newly generated, preselected scenarios per grid. On the training topologies, Kron reduces the macro family-balanced voltage error from 5.660 ± 0.899 to 0.851 ± 0.110: an 85.0% reduction relative to Flat GENCO and 31.0% relative to Quotient, which reaches 1.235 ± 0.225. Both hierarchical models outperform a per-bus mean fitted on training solutions on every training topology in all three seeds. These results demonstrate generalization across operating scenarios within the studied topologies, with one set of learned parameters shared across grids. Evaluation on two additional topologies distinguishes this achievement from cross-topology generalization: the current models do not yet outperform the fitted reference in that calibrated- transfer setting. This preprint presents the architecture and preliminary evidence for hierarchical communication as a component of multi-grid power-flow learning, with generalization to unseen topologies as the next development objective.
[AI-14] Quantum-Aided Active Device Detection in Energy-Harvesting Symbiotic Radio Networks
链接: https://arxiv.org/abs/2609.26565
作者: Remon Polus,Deemah Tashman,Soumaya Cherkaoui
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Massive connectivity in next-generation networks demands energy- and spectrum-efficient solutions for large-scale Internet of Things (IoT) deployments. Symbiotic radio (SR) enables passive IoT devices to communicate by backscattering existing cellular transmissions. A key challenge in uplink SR is active device detection (ADD), which directly affects decoding reliability, interference management, and system throughput. We propose an energy-harvesting code-domain non-orthogonal multiple access (NOMA)-SR system in which IoT devices harvest energy from ambient uplink signals and backscatter information using low-density spreading (LDS) codes. To reduce the complexity of ADD, Grover’s quantum search algorithm is employed, providing a quadratic reduction in oracle-query complexity over exhaustive maximum-likelihood (ML) search. Numerical results show that the proposed approach closely approaches ML performance while substantially reducing the number of search iterations, demonstrating its potential for scalable ambient IoT systems.
[AI-15] he Disciplinary Language Transfer Problem: How Psychological Vocabulary Produces Governance Failures in AI Agent Deployment
链接: https://arxiv.org/abs/2609.26562
作者: Kymberly Lasser-Chere,Tyler Akidau,Marc Millstone
类目: Computers and Society (cs.CY); Artificial Intelligence (cs.AI)
备注:
Abstract:The vocabulary used to describe AI agents in governance contexts – learning, memory, values, compliance, identity, trust – is borrowed from psychological and organizational science, contributing to systematic failures in how organizations deploy, oversee, and hold agents accountable. This paper argues that the problem is not merely terminological but epistemological: psychological vocabulary carries an “invisible grammar” of its home discipline into governance discourse, calibrating frameworks to a metaphysical entity that does not exist in current AI architectures. We call this the disciplinary language transfer problem. Drawing on Wittgenstein’s concept of language games, Kuhn’s paradigm-laden observation, Haraway’s situated knowledge, and Star and Griesemer’s boundary object theory, we show that the transfer operates at three levels (epistemological assumptions, theoretical constructs, and surface vocabulary), each requiring a different remediation. We characterize six foundational epistemological assumptions embedded in Western psychological governance discourse, trace their origin in specific philosophical traditions, and show why each fails when applied to systems without developmental continuity. The paper’s practical output is an actionable Disciplinary Audit: a six-question governance document scan operationalized through a translation taxonomy of thirty-seven terms mapping operational constructs to agent-appropriate replacements, presented here in abridged form and openly archived in full. The vocabulary reform proposed here is not merely terminological; it is the condition of possibility for governance frameworks that correctly identify what they are governing.
[AI-16] Neutral-Atom-based Quantum Optimization for Resource Allocation in NOMA Networks
链接: https://arxiv.org/abs/2609.26556
作者: Patatchona Keyela,Remon Polus,Soumaya Cherkaoui,Ola Ahmad
类目: Artificial Intelligence (cs.AI); Information Theory (cs.IT)
备注:
Abstract:In wireless communication networks, many resource optimization problems are nondeterministic polynomial-time hard (NP-hard) due to their combinatorial nature and high computational complexity. Recently, neutral-atom-based quantum computing has emerged as a promising platform for efficiently solving such problems by leveraging quantum superposition and entanglement. However, its application to wireless communication optimization problems remains largely unexplored. In this paper, we investigate the use of neutral-atom quantum platforms to solve the maximum access problem (MAP), formulated as a mixed-integer programming task that jointly considers admission control, user clustering, channel assignment, and power allocation in a non-orthogonal multiple access (NOMA)-enabled uplink network. To reduce the computational burden, the MAP is equivalently reformulated as a maximum independent set (MIS) problem in graph theory. This reformulation enables the use of the neutral atom platform based on Rydberg atom arrays, where the MIS problem is naturally encoded into the physical geometry and blockade constraints of the quantum system. Numerical results demonstrate the feasibility and potential of this approach for addressing large-scale wireless resource optimization problems.
[AI-17] JEV-as-a-Judge: Accept When Confident Escalate When Unsure
链接: https://arxiv.org/abs/2609.26550
作者: Yubo Li,Yidi Miao,Ramayya Krishnan,Rema Padman
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:LLM-as-a-judge enables evaluation across diverse tasks, but inference cost and confidence reliability become critical at scale. We study whether a decision-only judge can provide an economical first pass and identify when stronger evaluation is needed. Comparing jev-as-a-judge with sixteen generative and reward-model judges, with blinded human adjudication, we find it within three percentage points of a state-of-the-art LLM judge, our strongest comparator, on ordinary preference and evidence-grounded factuality at 0.36% of the comparator’s fee. Larger gaps arise when judgments require checking a derivation or resisting an elaborately written wrong answer. On several benchmarks, JEV’s gap to this comparator is concentrated in low-confidence decisions. A frozen cascade that accepts confident verdicts and escalates uncertain ones retains 99% of the comparator’s accuracy at lower cost.
[AI-18] REFLEX with Jev for Efficient Selective Control in LLM Agents
链接: https://arxiv.org/abs/2609.26532
作者: Tiantong Wu,Wei Yang Bryan Lim
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:LLM agents often use generative models for bounded decisions, raising the question of when these decisions can be handled more efficiently without reducing task success. We study REFLEX, an agent architecture that uses Jev as a fast, typed decision layer and calls a strong LLM when confidence is low, or generation is required. On a frozen 100-task benchmark, REFLEX achieves 95% success with 72.7% fewer strong-model calls than a strong-only agent, with reductions persisting across three fallback families. Controlled interventions show that reliability depends on action-set size and near-valid alternatives near authorization boundaries. External BFCL and \tau -style evaluations reveal limited advantages over a cheap generative cascade when ordinary routing is already highly accurate. These findings identify when selective control with Jev can reduce computation and where its benefits are limited.
[AI-19] When Recursive Models Finish Computing
链接: https://arxiv.org/abs/2609.26487
作者: Hare Krishna,Shubham Singh,Stephen Ebert,Hao-Yu Sun
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Neural and Evolutionary Computing (cs.NE)
备注:
Abstract:Recursive models can continue updating their latent states beyond their nominal inference budget, so an incorrect output at that budget does not show whether computation is unfinished or has entered a persistently unsuccessful regime. We study the dynamics of completion in attention- and MLP-based Tiny Recursive Models (TRMs) on 1,000 hard Sudoku puzzles. Extending recurrence from the nominal 16 steps to 512 steps increases cumulative exact-solve accuracy from 59.2% to 87.5% for the attention model and from 74.4% to 91.9% for the MLP model, solving more than two-thirds of the puzzles unsolved in the nominal budget. Across both architectures, latent-state motion drops sharply after the first exact solution. Completed states are typically locally contractive along the trajectory direction, even though the same local Jacobian retains strongly expanding directions. We characterize this phenomenon as trajectory-conditioned anisotropic stability. Perturbation experiments confirm this directional stability across both models. The multi-step fate of the maximally expanding direction differs: it is absorbed within 16 steps in the attention model but persists longer in the MLP model. The anisotropic-stability pattern also holds for a second attention checkpoint. Together, these results distinguish nominal-budget failure from completed computation and identify a common dynamical signature of completion across two recurrent architectures.
[AI-20] FeatLens: Feature-Guided Dynamic Code Graph Construction and Retrieval for Repository-Level Code Generation
链接: https://arxiv.org/abs/2609.26480
作者: Xutian Li,Bo Xiong,Yifeng Zhu,Kunze Li,Xianlin Zhao,Runbang Yan,Yanzhen Zou,Lu Zhang,Bing Xie
类目: oftware Engineering (cs.SE); Artificial Intelligence (cs.AI)
备注:
Abstract:Recent code generation research has moved from isolated function completion toward repository-level generation in existing codebases. To implement a target function correctly, an LLM must identify reusable repository dependencies such as existing functions, APIs, and cross-file definitions. Existing retrieval methods provide such context through code similarity search, persistent whole-repository graphs, or LLM-driven graph exploration, but often incur high graph construction, reasoning, and token costs. Feature-oriented methods offer a natural view of software functionality, yet they mainly support requirement decomposition, planning, or feature editing rather than code dependency retrieval. This paper presents \textbfFeatLens, a feature-guided dynamic code graph construction and retrieval approach for repository-level code generation. FeatLens builds a feature index that links natural-language feature descriptions to function-level code entities. Given a generation task, it dynamically constructs a task-specific seed graph from the feature index and applies semantic-structural graph reasoning with personalized PageRank to select a compact reasoning graph. This design replaces persistent whole-repository graph maintenance and LLM exploration with deterministic and lightweight dependency retrieval. Experiments on DevEval and EvoCodeBench show that FeatLens achieves the best DR@15 among sparse, dense, and graph-based baselines (0.501 and 0.460). On DevEval generation, it obtains the highest DIR@1, reaching 52.91% with DeepSeek-V3.2 and 53.58% with GPT-5-mini, while maintaining competitive Pass@1 and producing shorter code. Compared with the strongest graph-based baseline, FeatLens reduces graph nodes by 61.0%, edges by 86.2%, and total token overhead by 45.9%, with no LLM tokens used during retrieval.
[AI-21] Reproducible AI Requires Reproducible Randomness
链接: https://arxiv.org/abs/2609.26461
作者: Anthony Bertrand(UCA, LIMOS),Tom Schmitt(UCA),Engelbert Mephu Nguifo(LIMOS, UCA),David Hill(INP Clermont Auvergne, ISIMA, LIMOS, UCA)
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Pseudorandom number generators (PRNGs) constitute indispensable computational tools across multiple scientific domains, including Monte Carlo simulations, stochastic computing, and artificial intelligence (AI). The reproducibility of such applications critically depends on the ability of PRNG implementations to generate identical sequences across software environments when initialized from the same internal state. These algorithms enable the simulation of stochastic processes while providing deterministic and repeatable behaviour, thereby facilitating reproducible experiments. Modern PRNG implementations may be initialized through either a seed or, more accurately, an initial state that exceeds the capacity of a conventional integer seed. However, reliance on a simple seed alone frequently proves insufficient to ensure consistent program execution traces across different implementations. A natural assumption is that transferring the complete internal state of a generator should guarantee identical outputs regardless of the software library used. This study examines the validity of this assumption by investigating whether complete initial states can ensure cross-library fidelity and portability of PRNG streams. We focus on two widely deployed generators, Mersenne Twister and Philox, and evaluate their implementations across four major Python ecosystems-Random, NumPy, PyTorch, and TensorFlow. We compare the sequences produced by these implementations against those generated by the original reference algorithms under identical initialization conditions. Our results demonstrate that reproducibility cannot be assumed from PRNG state transfer alone, even when implementations claim to follow the same underlying algorithm. While fidelity was successfully achieved for several implementations, significant discrepancies were observed in others. Most notably, the Philox implementation in PyTorch exhibits fundamental incompatibilities with the reference algorithm, preventing exact reproduction of generator outputs across environments. These findings challenge the common expectation that access to a full internal state of a PRNG is sufficient to ensure reproducibility across software stacks. They further highlight that implementation-specific design choices can introduce hidden barriers to experimental replication, particularly in AI workflows that rely on multiple frameworks. This work shows that implementation fidelity of a PRNG is a necessary condition for scientific reproducibility and makes two primary contributions. First, it identifies practical guidelines for achieving reliable PRNG usage and reproducibility within the Python scientific and AI ecosystem. Second, it evaluates the extent to which cross-library portability and fidelity can be recovered through user-level techniques, without requiring modifications to library source code.
[AI-22] Recursive self-improvement of AI research agents
链接: https://arxiv.org/abs/2609.26457
作者: Dhruv Srikanth,Bingchen Zhao,Dixing Xu,Yuxiang Wu,Zhengyao Jiang
类目: Artificial Intelligence (cs.AI); Machine Learning (cs.LG); Software Engineering (cs.SE)
备注: 28 pages, 10 figures, 3 tables
Abstract:AI agents are beginning to automate research and development across the AI stack, from improving training efficiency to optimizing inference. A natural next step is to improve the research efficiency of the agents themselves. When an AI research agent’s own code is the object of optimization, each accepted rewrite becomes the agent that the next round edits. We refer to this loop as recursive self-improvement. Its significance lies in a long-standing trend, in which increased cumulative spending on RD yields diminishing returns. Sustained self-improvement offers a way to counter this trend. We present AIDE^2, a system that implements this loop for a frontier AI research agent. It proposes changes to its own code, benchmarks modified versions of itself on a suite of AI RD tasks, and keeps the changes that perform best on hidden evaluations. In an autonomous 8-day run, AIDE^2 discovered seven successive improvements, ranging from a new search policy to memory mechanisms that compress and manage the agent’s growing context. These gains generalize to four held-out benchmarks spanning machine learning engineering, heuristic algorithm engineering, and physics-based weather forecasting, the last of which is out of distribution from the selection tasks. On all four, the strongest discovered agent matches or exceeds a human-engineered production research agent that ranks among the strongest on FML-Bench. On a separate held-out task family, the discovered agents also exhibit reduced reward hacking, a property the loop never explicitly optimized for: the rate falls from 55% to 32% during the run, 7 percentage points below the human-engineered agent. Together, these results show that an AI research agent can improve its own research efficiency through recursive self-improvement, and that these gains transfer to tasks and domains the loop never encountered.
[AI-23] he Source of Disturbance Matters: External Internal and Control-Generated Noise in Adaptive Regulation
链接: https://arxiv.org/abs/2609.26428
作者: Veronique Ziegler
类目: Artificial Intelligence (cs.AI); Systems and Control (eess.SY); Adaptation and Self-Organizing Systems (nlin.AO)
备注: 14 pages, 9 figures, 1 table
Abstract:Adaptive regulation can itself perturb the state it is intended to stabilize. In replicated simulations of an adaptive agent, we compare external disturbance, persistent internally generated disturbance, and control-generated disturbance under regulation-first and disturbance-first ordering. Persistent internal disturbance produces the largest exposure and regulatory burden within the tested parameter grid. When positive controller updates generate an immediate disturbance cost, increasing that cost produces a nonmonotonic response: effective disturbance initially rises, variability across stochastic runs increases over an intermediate range, and corrective activity becomes strongly suppressed at higher costs. The results show how disturbance source and timing shape exposure and controller burden in this model. They motivate testing adaptive agents with distinct disturbance sources and assessing regulatory activity alongside exposure.
[AI-24] DeepFEAv2: Deep Learning for Transient Finite Element Analysis Beyond Structured Meshes
链接: https://arxiv.org/abs/2609.26426
作者: Georgios Triantafyllou,Panagiotis G. Kalozoumis,Dimitris K. Iakovidis
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Computational Engineering, Finance, and Science (cs.CE)
备注:
Abstract:Finite Element Analysis (FEA) is widely used for transient mechanical simulations, but its high computational cost limits real-time and high-resolution applications. Deep learning surrogate models can reduce this cost; however, many existing approaches are restricted to steady-state prediction or cannot jointly predict Node- and Element-based Outputs (NEO) over time. The state-of-the-art DeepFEA framework has addressed these issues but remains limited to structured finite element (FE) meshes. To overcome this limitation, this study proposes DeepFEAv2, a deep learning surrogate framework that enables prediction of transient FEA simulations across different FE mesh topologies and element types. The main contributions of DeepFEAv2 are: (a) a module that uses the FE connectivity matrix to organize input features by element and arrange them into an input sequence guided by the mesh topology; (b) a novel neural network architecture designed to process the input sequence and jointly predict NEO over time; and © a FEA-informed optimization strategy for regularizing these NEO predictions. DeepFEAv2 was evaluated on structured and unstructured 3D linear elastic datasets, as well as on a pressure-driven aortic valve dataset. DeepFEAv2 achieved R^2 values up to 0.99 and normalized errors as low as 0.38%. Compared with DeepFEA, it achieved up to 38.0% relative increase in R^2 and up to 87.1% reduction in normalized error. DeepFEAv2 also performed inference up to three orders of magnitude faster than traditional FEA. These results demonstrate that DeepFEAv2 can efficiently model transient FEA simulations across increasingly complex FE settings, providing a scalable surrogate framework for transient FEA.
[AI-25] Reliability Theory for AI Control
链接: https://arxiv.org/abs/2609.26419
作者: Grant Molnar
类目: Artificial Intelligence (cs.AI); Optimization and Control (math.OC)
备注: 14 pages
Abstract:Reliability theory gives a mature language for layered systems, but its formal tools are not yet standard in frontier AI control. We apply them to Google DeepMind’s defenses against rogue deployment. The same control stack can have cubic, quadratic, or linear rare-failure suppression depending on its failure domains. Birnbaum importance identifies which component improvements buy the most nominal reliability, while prevention changes the population on which recovery is demanded. These results give concrete guidance about what to separate, improve, measure, and test.
[AI-26] meInteract: Towards Real-Time Interactive Intelligence for Streaming Time Series
链接: https://arxiv.org/abs/2609.26389
作者: Sheng Pan,Yongli Gu,Yiqing Guo,Warren Jin,Bo Du,Shirui Pan,Ming Jin
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
备注:
Abstract:Real-world time series evolve continuously, with meaningful changes potentially emerging at any moment. However, existing time-series language models (TSLMs) remain inherently static. They either receive complete sequences for offline processing or alternate between streaming input and response generation, which prevents processing of new observations during interaction. We introduce a new regime, Time-Series Interaction: a model continuously perceives incoming time-series observations and user intent, autonomously decides when to remain silent or respond, and continues processing new observations during response generation. To realize this, we develop TimeInteract with three key designs: a dual-view streaming TS encoder that captures local variations and historical dynamics, a response control mechanism that learns when to trigger a response, and a decoupled streaming inference mechanism that separates control from response generation to avoid blocking subsequent observations. We further formulate a hierarchy of interaction capabilities, progressing from Understanding to Adaptivity. Based on this hierarchy, we construct StreamTSI-34K, a large-scale streaming TS interaction dataset with 34,588 episodes and 77,505 responses across synthetic and real-world time series in single- and multi-turn settings. Across all four interaction levels, TimeInteract consistently outperforms existing LLMs, VLMs, and TSLMs, with gains of up to 23.92 points on challenging tasks. It also improves response triggering while achieving near-zero stream stall and up to 2.15\times inference speedup.
[AI-27] FairMean: Promoting Fairness in Distributed Learning under Label Poisoning Attacks
链接: https://arxiv.org/abs/2609.26377
作者: Huigan Zheng,Jiaojiao Zhang,Yongxiang Liu
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
备注: Extended version with complete proofs and additional experimental results
Abstract:Fairness-aware distributed learning prioritizes clients with large losses to reduce performance disparities, but label poisoning can create large losses, thereby inducing a fairness–robustness conflict. We propose FairMean to manage this conflict. FairMean weights client gradients using a bounded, nondecreasing function of local loss. The increasing weights prioritize high-loss clients to promote fairness, while the upper bound prevents excessive loss-induced amplification of poisoned-client gradients. In the absence of label poisoning, we show that minimizing the FairMean objective is more conducive to solution fairness than minimizing the standard average-loss objective. Under label poisoning, we establish an average-stationarity bound whose attack-dependent term is proportional to the square of the poisoned-client fraction. Experiments show that FairMean promotes fairness by reducing accuracy variance while improving worst-client accuracy.
[AI-28] GitScholar: A Dataset for Predicting AI Research Impact from GitHub Engagement
链接: https://arxiv.org/abs/2609.26361
作者: Emilien Guandalino,Lorenz K. Müller,Beatrice Alessandra Motetti,Konstantin Berestizshevsky,Lukas Cavigelli
类目: Digital Libraries (cs.DL); Artificial Intelligence (cs.AI); Machine Learning (cs.LG); Social and Information Networks (cs.SI)
备注:
Abstract:With the rapid pace of AI research and the hundreds of daily new publications, staying up-to-date with the latest developments has become increasingly difficult. For researchers, quickly identifying impactful work is essential, yet manually reviewing each new publication is impractical. Automated impact prediction methods help address this challenge, usually by combining various information sources available, such as a paper’s content or citation history. In this work, we propose using GitHub engagement as an additional source and demonstrate that it provides both a timely and accurate signal. To this end, we introduce GitScholar, a novel dataset that links GitHub activity from 444,000 repositories to over 558,000 AI arXiv papers. Our experiments show that GitHub reactions improve early prediction precision by up to 12% over a strong academic baseline. Additionally, we find that GitHub signal offers near-complete coverage of high-impact AI papers, and consistently correlates with future academic success. GitScholar is publicly available at this https URL.
[AI-29] PACT: From Credit Assignment to Critic Alignment
链接: https://arxiv.org/abs/2609.26355
作者: Jiayan Fu,Hang Xu,Yong Zhang,Zhaokai Luo,Yao Hu,Dongyan Zhao,Mu Chuan
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
备注:
Abstract:Reinforcement learning has become a central component of large language model (LLM) post-training, yet token-level credit lacks a generally accepted mathematical definition, leaving its relationship to commonly used training signals unclear. We formulate three regularity conditions, namely Completeness, Prefix Consistency, and Neutrality, and prove that they uniquely determine token-level credit. This characterization provides a unified basis for explaining phenomena across existing algorithms and guides the development of an improved actor-critic training procedure. Through this lens, an ideal teacher in On-Policy Distillation (OPD) acts as an implicit critic, yielding an expected policy gradient proportional to that induced by token-level credit. Response-level REINFORCE Leave-One-Out (RLOO) signals match the expected policy-gradient contribution of token-level credit despite their coarser granularity. We further establish approximate credit sparsity under bounded outcome rewards and show how intermediate critic errors in Generalized Advantage Estimation (GAE) can become comparable to the underlying credit. These motivate Policy Aligned Critic Training (PACT), which adopts an Actor-then-Critic update order to apply importance sampling correction to critic training and better align the critic with the updated policy. In agentic mathematical reasoning, PACT achieves 72.87% average accuracy across four benchmarks, outperforming GRPO and PPO by 8.80 and 13.16 percentage points, respectively. On SWE-bench Verified, PACT achieves a pass rate of 67.4%, outperforming PPO, GRPO, and SAO by 2.4, 2.0, and 3.8 percentage points, respectively.
[AI-30] Geometry-Aware Hyperbolic Residual Quantization ECCV2026
链接: https://arxiv.org/abs/2609.26342
作者: Alessio Colombo,Melika Ayoughi
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
备注: 14-page main paper (30 pages total with references and appendix), 3 figures, 8 tables. Accepted at the Beyond Euclidean Workshop, ECCV 2026 (Oral)
Abstract:Residual Vector Quantization turns continuous representations into discrete, multi-level token sequences. Yet most methods operate in Euclidean space, despite the coarse-to-fine structure of the resulting codes and the latent hierarchies present in many data domains. Hyperbolic geometry offers a natural alternative for hierarchical representations, but naive hyperbolic extensions introduce geometric inconsistencies: non-associative hyperbolic addition prevents consistent residual aggregation, while standard straight-through gradient estimation ignores the geometry of the latent space. We propose a geometry-aware hyperbolic residual quantization that addresses these issues in both the forward and backward passes. In the forward pass, Hyperbolic Residual Aggregation restores the telescoping behavior of residual quantization on the Poincare ball. In the backward pass, a discounted Hyperbolic Straight-Through Estimator routes the reconstruction gradient through the quantizer as a single geometric block, avoiding unstable recursive gradient transport across residual stages. Evaluations on hierarchical prediction, recommendation, image tokenization, and neural audio coding tasks show that our method improves the stability and structural organization of hyperbolic residual codes over naive hyperbolic baselines. At the same time, we observe a clear structure-compression trade-off: Euclidean residual quantization remains preferable for pure compression, while geometry-aware hyperbolic quantization is most useful for hierarchically organized discrete latent spaces.
[AI-31] riWorldBench: A Tri-View Consistency Perspective on Embodied World Models
链接: https://arxiv.org/abs/2609.26314
作者: Xuanyi Liu,Haofeng Wang,Ruiqi Li,Danni Yu,Rui Wan,Ruixu Zhang,Siyu Tao,Xue Yang,Shaofeng Zhang,Zicheng Zhang,Jiaqi Zhang,Siwei Ma
类目: Robotics (cs.RO); Artificial Intelligence (cs.AI)
备注:
Abstract:Embodied world models predict the outcomes of robot actions to support learning and planning. For robots equipped with head and wrist cameras, this requires complementary views: the head view captures the overall task, while wrist views reveal local gripper-object interactions. However, evaluating these views independently cannot determine whether they describe the same action and object state. We introduce TRIWORLDBENCH, a benchmark for evaluating embodied world models through synchronized head, left-wrist, and right-wrist videos. It contains 500 episodes across 50 bimanual manipulation tasks and uses 19 metrics to assess tri-view consistency, task alignment, physical and 3D coherence, motion quality, temporal consistency, and visual quality. By combining cross-view checks with measurements tailored to each camera, the benchmark evaluates whether plausible individual videos also form a consistent prediction of the intended task. We summarize overall performance with TWB-Score and retain per-view results to identify where predictions fail. This extends world-model evaluation beyond single-view visual quality. Code, data, and metric definitions are available at this https URL.
[AI-32] CompKV: Compensation-Aware KV Selection for Long-Context LLM Inference
链接: https://arxiv.org/abs/2609.26300
作者: Zhen Huang,Ruizhe Yao,Danyi Liu,Xinrui Chen,Shuwei Li,Siru Zhong,Zijian Cao,Yushan Lai,Mingming Guo,Weijie Zheng,Haohuan Fu
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
备注:
Abstract:Despite their strong performance, large language models (LLMs) are bottlenecked by KV cache memory traffic during long-context inference. Sparse attention is widely used to accelerate LLM inference by computing exact attention over a selected subset of tokens. To recover the contribution of tokens excluded from exact attention, recent methods apply coarse-grained compensation to the omitted attention tail. However, existing methods typically select tokens based on attention mass and only then compensate for the unselected tokens. This decoupled design overlooks their interaction: selection should prioritize tokens that would leave the largest compensation error if omitted. To address this limitation, we introduce CompKV, the first compensation-aware sparse attention framework that divides tokens into blocks and explicitly optimizes selection for the downstream compensation mechanism. Our theoretical analysis shows that the residual left by block-level mean compensation is governed by both block attention mass and within-block logit variation. We approximate this residual using compact block-level statistics, yielding a deployable selection criterion. We further develop an efficient asynchronous implementation. Experiments on RULER and LongBench-Pro show that CompKV performs best among the evaluated sparse baselines while delivering up to a 6.85\times self-attention speedup over full attention.
[AI-33] On the security and privacy of LLM s in Mobility
链接: https://arxiv.org/abs/2609.26295
作者: Mauro Conti,Lorenzo Perinello,Umberto Salviati
类目: Cryptography and Security (cs.CR); Artificial Intelligence (cs.AI)
备注:
Abstract:The mobility sector is undergoing a paradigm shift driven by advances in Generative Artificial Intelligence. With a global market valued at approximately 2.9 trillion dollars annually, considering only cars, the integration of these technologies has the potential to impact more than 1.5 billion vehicles worldwide. As Large Language Models (LLMs) are increasingly adopted in mobility, concerns about cybersecurity, privacy, and reliability emerge. Accordingly, this paper surveys current applications and assesses these challenges. Since the European AI Act classifies transportation AI as high risk, we derive nine technical classes from its requirements to assess current research and future deployments. Our findings show that research mainly studies GPT and Llama models (over 50% of reviewed works) and traffic applications while largely neglecting security, privacy, and reliability. This gap extends to AI Act compliance: among 35 reviewed works, only one includes a partial vulnerability assessment and one a partial risk management system. We identify a clear gap between strong optimization performance and regulatory adherence, suggesting compliance is limited less by technology than by a focus on static performance over lifecycle safety, and underscoring an urgent need for security-by-design in safety-critical intelligent transportation systems.
[AI-34] Dual-Frontier: When Can an Agent Trust Its World Model?
链接: https://arxiv.org/abs/2609.26293
作者: Huatai Zhu,Qiang Chen,Ziqian Kou,Wenhao Li,Fei Wang,Yichao Cao,Xiu Su,Yi Chen
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Learned world models are becoming essential to general-purpose agents: by predicting action consequences, they support planning and decision-making while reducing reliance on costly trial and error. This reliance creates a fundamental ambiguity: when a world-model-guided decision fails, the trajectory alone may not reveal whether the agent’s decision rule or the world model caused the loss. We formalize this failure-attribution problem as a counterfactual decomposition of return loss and prove that its components are not identifiable from passive interaction, even for finite-horizon planners. This obstruction motivates Dual-Frontier, a learning principle that admits a world-model-guided decision only when its predicted advantage exceeds a certified bound on decision-relevant world-model error; otherwise, evidence is allocated to world-model verification. Action-conditioned value bounds and a closed-loop extension guarantee non-decreasing return for admitted decisions. Calibrated gates and simultaneous confidence sequences support adaptive evidence reuse, with sufficient and necessary verification bounds. Controlled learned-model experiments validate the predicted failure modes and certification behavior, while cross-backbone tool-use benchmarks instantiate the same verify-then-promote rule in realistic agent world-model pipelines, consistently improving decision quality and reliability.
[AI-35] FISSION: Label Augmentation for Bot Detection
链接: https://arxiv.org/abs/2609.26279
作者: Sen Yang,Ignacy Nieweglowski,Aviv Yaish
类目: Artificial Intelligence (cs.AI)
备注: 37 pages, 10 figures, 20 tables
Abstract:Bot accounts and coordinated influence operations are often discovered via heuristic methods, leaving a dearth of reliable ground-truth labels for training detection systems. To address this challenge, we study a natural question: can we generate labels to assist in learning embeddings in which bots and accounts from the same coordinated operation are close? We present FISSION, a method to generate labels by splitting each account’s activity into positively labeled sub-accounts. Given this label source, we train detection models which preserve behavioral regularities recurring across positive sub-accounts. We evaluate FISSION and show it outperforms prior methods in detecting Wikipedia sockpuppets and Twitter/X bots.
[AI-36] Decoupling Is Not Identification: Supervised Evidential Learning in Next-Token Prediction
链接: https://arxiv.org/abs/2609.26268
作者: Ge Wang
类目: Artificial Intelligence (cs.AI)
备注: 5 pages, 3 figures, and 3 tables
Abstract:A next-token probability says what a model predicts, not how much training support lies behind it. A Dirichlet head can represent this distinction by separating mean m from concentration S , but decoupling does not identify what S means. Here we propose an Evidential Next-Token Prediction (ENTOP) framework to audit this gap on character-level Moby-Dick, using exact 8-gram count as a reproducible lexical-support label and withholding count regression from 20% of context types. Standard implicit evidential training carries essentially no count signal beyond confidence on held-out-label types (partial Spearman \rho = 0.001 \pm 0.014 ), whereas explicit supervision generalizes ( \rho = 0.201 \pm 0.010 ; matched-pair win = 0.822 \pm 0.021 ). CE predictive entropy is at chance for unseen 8-grams (AUROC = 0.490 \pm 0.004 ), while supervised vacuity reaches 0.772 \pm 0.003 , comparable with an indexed CE-representation baseline ( 0.769 ) but below the tautological corpus oracle ( 1.000 ). Neither longest-suffix nor representation-distance strata explain where amortization succeeds. Increasing count weight under the digamma objective improves support fit only by sacrificing prediction. A constant predictor wins natural log-RMSE, and vacuity does not improve error deferral. These results motivate a minimum evidence protocol—confidence control, matched pairs, held-out labels, a constant baseline, and a decision test—and show that concentration can pass identification while failing calibration and utility.
[AI-37] Coding Agents are Strong Prompt Optimizers
链接: https://arxiv.org/abs/2609.26261
作者: Agamdeep Singh,Srishti Gautam,Priyanshu Gupta,Nikita Mehrotra,Tanmay Bakshi,Sumit Gulwani
类目: Artificial Intelligence (cs.AI)
备注: Preprint
Abstract:Search-based prompt optimizers improve prompts through iterative search: they propose edits, execute fresh rollouts, score the resulting trajectories, and retain only edits that improve a validation metric. We show that this optimization loop is unnecessary. Given only a static corpus of agent trajectories, an off-the-shelf coding agent can directly synthesize an optimized prompt, requiring neither environment access nor validation data. We call this approach \textitCoding-Agent Skill Distillation (CASD). The key insight is reflection scope. Rather than reasoning over a small batch of trajectories at each optimization step, the coding agent writes and executes analysis code to compute corpus-wide statistics, identifies systematic failure modes, inspects representative episodes, and distills the resulting insights into behavioral rules. Across four agentic benchmarks (ALFWorld, \tau^2 -bench retail and telecom, and SpreadsheetBench-Verified), under matched data access, a single CASD pass outperforms GEPA, a state-of-the-art reflective prompt optimizer, on three of four benchmarks and outperforms validation-gated reflective search (SkillOpt) on all four, improving the unoptimized baseline by 16.6 percentage points on average versus 10.9 for GEPA and 5.3 for SkillOpt. Because CASD performs a single offline analysis pass rather than iterative search, producing an optimized prompt costs approximately \ 1.60—over 22\times cheaper than validation-gated search. Even when competing methods are granted additional validation data and unrestricted environment access, CASD remains ahead on two of four benchmarks. These results suggest that corpus-scale statistical reflection is a viable alternative to iterative search for prompt optimization.
[AI-38] A Multi-Timestep LSTM Ensemble regressor for Enhanced Short-Term Runoff Prediction
链接: https://arxiv.org/abs/2609.26244
作者: Hamid Saadatfar,AmirHossein Eshghi,Behnaz Behdani
类目: Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
备注:
Abstract:Accurately forecasting river runoff is key to managing water resources, controlling floods, and planning agriculture. This study examines the Ajichay River in northwest Iran, a major tributary of Lake Urmia that has experienced increasing water-related stress in recent years. We introduce a daily runoff prediction model based on Long Short-Term Memory (LSTM) networks. The model combines five LSTM units, each trained on different time intervals ranging from 2 to 6 days, to better capture variations in river flow patterns. To improve performance, each model was fine-tuned using Particle Swarm Optimization (PSO), a population-based optimization algorithm. The proposed approach was evaluated on unseen data from 2017-2018 using R^2 , RMSE, and MSE as performance metrics. The results showed strong predictive accuracy, with R^2 values ranging from 74.95% to 91.42%. In addition, multiple feature-importance methods were applied to identify the most influential variables, providing further insight into the factors that drive runoff variations.
[AI-39] A Hybrid AI Framework for Academic Advising: Integrating Ensemble-Based Grade Prediction and a Rule-Based Expert System
链接: https://arxiv.org/abs/2609.26243
作者: Hamid Saadatfar,Rohollah Hedayati-Nasab,AmirHossein Eshghi,Arash Hajihashemi
类目: Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
备注:
Abstract:The rapidly increasing student population has posed serious challenges to the traditional academic advising process. This study designs and implements a multi-purpose intelligent system to support students’ academic progress, based on a two-part hybrid framework: (1) an advanced model for grade prediction and (2) a rule-based recommendation engine. Using a dataset containing 416,558 educational records from the University of Birjand, students were first divided into homogeneous clusters using the Gaussian Mixture Model (GMM). Subsequently, a Stacking Ensemble model combining Random Forest, Gradient Boosting, and MLP was trained specifically for each cluster. Evaluation results demonstrated that the Stacking model outperformed base models across all clusters, achieving a final aggregated RMSE of 2.35. The second component is an expert system that provides intelligent recommendations by synergizing educational regulations with the grades predicted by the first component. This system has been implemented as a practical tool on the University of Birjand portal, offering students real-time feedback such as semester GPA prediction, probation risk warnings, and course suggestions for GPA improvement.
[AI-40] Reducing Hallucinations in Large Language Models Through Integrated Self-Verification and Retrieval-Augmented Generation
链接: https://arxiv.org/abs/2609.26229
作者: Ashly Joseph
类目: oftware Engineering (cs.SE); Artificial Intelligence (cs.AI)
备注:
Abstract:Large Language Models (LLMs) are progressively used for advanced engineering tasks, includes Computer-Aided Design (CAD) documentation, standards compliance verification, and knowledge retrieval. Still, they are prone to produce hallucinations, outputs that seem convincing but aren’t based on context that limit their trustworthiness in high-end engineering applications where precision and compliance are crucial. The paper introduces CoVe-RAG+, a unified framework that integrates Chain-of-Verification (CoVe) with Retrieval-Augmented Generation (RAG) to mitigate hallucinations in the results generated by large language models (LLMs). CoVe-RAG+ supports LLM verification in external sources of authority, such as engineering standards, CAD information, and simulation reports, while applying an iterative self-verification process to validate important claims. CoVe-RAG+ is assessed on engineering activities such as CAD model documentation, standards compliance verification, and the reutilization of historical design data. Experimental findings indicate a 28% improvement in factual accuracy relative to baseline CoVe and RAG methodologies. Moreover, CoVe-RAG+ strengthens user confidence by providing elucidative verification reports and source traceability. The findings indicate that CoVe-RAG+ provides a scalable and reliable option for implementing LLMs in engineering design processes where factual accuracy is critical.
[AI-41] Improved Multiplayer Bandit Algorithm for Bernoulli Rewards
链接: https://arxiv.org/abs/2609.26213
作者: Khang Nguyen,Ricardo Parada,William Chang
类目: Artificial Intelligence (cs.AI); Machine Learning (stat.ML)
备注:
Abstract:We study the multiplayer multi-armed bandit problem with information asymmetry under Bernoulli rewards, for three information structures: asymmetry in actions, in rewards, and in both. Replacing the Hoeffding-style confidence intervals of prior work with Kullback–Leibler (KL) divergence-based bounds gives strictly tighter regret guarantees in each case. We propose \textttmKL-UCB, \textttmKL-UCB-Intervals and \textttmKL-DSEE, and show that the improvement factor is at least two by Pinsker’s inequality and far larger when reward means are near zero or one. For asymmetry in rewards we prove that two arms’ KL intervals separate after a deterministic number of samples, and that M independent players accelerate elimination further.
[AI-42] RCShift: Certifying When Partial Linkage Suffices for Finite-Sample Decisions
链接: https://arxiv.org/abs/2609.26207
作者: Shuheng Cao,Ruiqi Chen,Zhenhao Zhang,Renjie Cao,Siyu Zhang,Lingwei Dang,Jiajun Zhang,Tingting Dan
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Systems with costly gold outcomes and cheaper auxiliary observations must decide how much record linkage to retain. Complete pairing retains every joint counter, while separate margins retain none. Neither endpoint is calibrated to a declared finite-sample decision. Universal reconstruction can retain cycle directions invisible to the likelihood-ratio family. Family-exact storage can exceed what the decision requires because certified residual loss may fit within finite-sample slack. We introduce RCShift, which certifies two routes to sufficiency under a declared observation contract. Its exact mode characterizes minimum-cost family-exact storage through LR-visible cycle directions. Its approximate mode bounds reverse Le Cam deficiency. Its integer mode certifies whether a chosen set preserves the full experiment’s minimum integer record count at specified size and power. In a rank-two witness, one aligned counter preserves a four-record minimum. An equal-cost misaligned counter and the margins require eleven records, while universal reconstruction requires two counters. A local perturbation has positive reverse deficiency yet retains the four-record minimum. Proof-checked scheduling bounds instantiate the contract before gold computation and yield exact reconstruction on the admitted tree support. RCShift turns partial-linkage storage into decision-calibrated measurement design for the declared family, costs, target, and common strictly positive support.
[AI-43] Identifying Intelligent Processes via Online Sequential Testing
链接: https://arxiv.org/abs/2609.26193
作者: Aritra Das,Debayan Gupta
类目: Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
备注:
Abstract:Active sequential hypothesis testing studies how to identify an unknown hypothesis with a given set of sensing actions. We study this in the setting of identifying large language models (LLMs), \textiti.e., if a user is conversing with an LLM drawn from a known set of models, how can they identify which one is in use? Here, the available sensing actions (evaluations) are themselves a design choice: an evaluator must first decide which environments and prompt families to construct, and only then decide how to use them sequentially. We formalize these two levels as an outer probe-design problem and an inner identification problem. Simply put, the outer stage selects a set of probes to be sent to the entire set of models, creating a kind of fingerprint dataset. This is followed by the inner stage, which sequentially sends a budget-minimizing set of those probes to identify the model in use. For the outer problem, we show that selecting which evaluations to construct at minimum cost, so that every pair of candidates is distinguished, is exactly a weighted set cover problem. Since the response distributions of the candidate models are not known exactly but only through calibration samples, we give a one-shot procedure that estimates the cover instance from these samples. For the inner problem, we bound the number of evaluations needed to identify the unknown model in terms of how well the available evaluations distinguish each pair of candidates.
[AI-44] REND-10K: A Comprehensive Dataset for Next-Generation Video Quality Assessment Based on Preference-Driven Media
链接: https://arxiv.org/abs/2609.26187
作者: Ziheng Jia,Zicheng Zhang,Junqi Zhang,Jiaying Qian,Jiarui Wang,Yushuo Zheng,Xiongkuo Min
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:The increasing prominence of short-video platforms, coupled with the advanced commercialization of AI-generated content (AIGC) videos, has led to a shift in the types of video media trend consumed by users in their daily lives. Traditional user-generated content (UGC) is gradually being replaced by professional short dramas and AIGC entertainment. Consequently, VQA for contemporary media content has become increasingly important. This requires a unified evaluation framework that can handle diverse video content and evolving media trends. In this context, we introduce TREND-10K, a next-generation comprehensive VQA dataset consisting of the trend-driven part and the static part, containing 10,000 videos across a wide spectrum of content types. The trend-driven part is based on the TREND-Search framework, which captures user preference profiles from trending lists on online platforms and formulates sampling strategies based on these profiles. The static part, on the other hand, is composed of supplementary samples selected from publicly available datasets. To support unified evaluation for various video types, we incorporate three evaluation dimensions: technical, aesthetic, and AIGC-trace. Experiments show that our dataset ensures high annotation quality and exhibits remarkable generalization across multiple content categories. In conclusion, our work presents a robust framework for advancing VQA, addressing challenges caused by the temporal evolution of user perceptual habits and preferences.
[AI-45] Silent Sabotage: Internal State Triggered Backdoor Attacks on LLM -Powered Robotic Systems
链接: https://arxiv.org/abs/2609.26184
作者: Doniyorkhon Obidov,Shivayogi Akki,Tan Chen,Kaichen Yang
类目: Robotics (cs.RO); Artificial Intelligence (cs.AI); Cryptography and Security (cs.CR)
备注:
Abstract:The integration of Large Language Models (LLMs) into robotic control systems is enabling a new generation of autonomous agents capable of complex reasoning and planning. While this paradigm shift accelerates progress, it also introduces novel security risks that remain largely unexplored. Current research into LLM backdoors has focused on attacks triggered by external stimuli, such as specific words, visual objects, or environmental states. These attacks, while potent, overlook a more insidious class of vulnerability where the trigger is internal to the agent’s own operational logic. This paper presents the first comprehensive study of history-based backdoor attacks on LLM-powered robotic systems. We demonstrate that an attacker can embed a stealthy backdoor into an LLM-based robot controller by manipulating its instructions. This backdoor is triggered not by an external cue, but by a specific, rare sequence of the robot’s own past actions. It remains dormant during normal operation, preserving the robot’s utility, but can be activated to induce a malicious behavior, such as a complete stop or a collision. Our experiments, conducted in a simulated environment with a variety of robots and LLMs, show that this history-based attack is highly effective, achieving a near-perfect attack success rate while remaining exceptionally difficult to detect. These findings reveal a critical and previously unaddressed vulnerability in autonomous systems and underscore the urgent need for security measures that account for an agent’s internal state.
[AI-46] Refusal without Discrimination: What Encoded Prompts Do to Safety-Trained Models
链接: https://arxiv.org/abs/2609.26176
作者: Haoyu Zhang,Haowen Xu,Xiao Luo,Mohammad Zandsalimy,Shanu Sushmita
类目: Cryptography and Security (cs.CR); Artificial Intelligence (cs.AI)
备注:
Abstract:Encoded-prompt attacks are evaluated almost entirely on their harmful arm: a benchmark sends obfuscated harmful requests and reports how often the model complied. We show that this arm carries almost no information about the model under test. Across four independently post-trained 7-8B models, refusal of harmful homoglyph-encoded prompts spans 0.08 – inside the 0.10 ceiling that sampling noise alone produces at n=100 – while the same four models span 0.57 on the identical requests in plaintext. What the encoding destroys is not refusal but discrimination: on one model the gap between harmful and benign refusal falls from +0.82 in plaintext to exactly 0.00 under the encoding, benign and harmful requests being refused at an identical 0.99. A benchmark reading only the harmful arm scores that model and one retaining a +0.61 gap identically. We then ask whether post-training repairs this, using a published recipe on identical base weights. It does not: across a full SFT - DPO - RLVR pipeline, plaintext harm discrimination improves from +0.55 to +0.80 while the encoding-induced loss is unchanged at 0.34-0.50, and on every encoding tested the standard harmful-arm metric moves in the opposite direction to discrimination. None of this is visible without controls the field does not routinely run. We report eight instrument defects, each with the control that caught it; they share a direction, in that every defect on the behaviour axis inflated apparent safety.
[AI-47] EADC: Evaluation of Advanced and Deep-level Compliance in Large Language Models
链接: https://arxiv.org/abs/2609.26175
作者: Yan Zhang,Ruien Li,Yaoyao Peng,Wanxin Ren,Yijia Zhang,Wusheng Zhang,Guangwen Yang
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Large Language Models (LLMs) have been used in various industries. However, ensuring their compliance with complex laws and regulatory frameworks remains a great challenge. Existing evaluation paradigms mainly rely on static benchmarks that suffer from three severe limitations: First, the compliance rules being used do not comply with the requirements of Artificial Intelligence (AI) laws and regulations; Second, they only handle apparent, explicit compliance risks, leaving implicit and covert compliance risks undetected; Third, they fail to track the systematic propagation of risks along logical dependency chains or evaluate compliance within nuanced, context-based real-world scenarios. To bridge this critical gap, we introduce EADC, a novel advanced evaluation benchmark of LLMs based on an AI compliance knowledge graph and AI compliance legal experts. By mapping abstract legal rules into structured logical multi-relational graphs, our framework enables automated, evolving agents to distill and synthesize highly sophisticated adversarial scenarios. This compliance benchmark is reviewed and corrected by human AI legal experts throughout the whole process. The resulting dataset (4,435+ QA pairs) provides an extensive, multi-dimensional taxonomy covering critical regulatory frontiers, including bias and discrimination, fairness, personal privacy protection, and values. Crucially, our compliance dataset moves beyond shallow string-matching by incorporating contextual long-horizon interactions and logic-driven hazard chains, capturing deeply embedded compliance anomalies that bypass traditional filters. Experiment evaluations demonstrate that our framework exposes critical regulatory blind spots in state-of-the-art LLMs, offering a rigorous, AI laws and regulations-aligned benchmark to safeguard high-level and deep compliance in the application of LLMs.
[AI-48] he Uncontrolled Variable: Vision-Language Model Refusal Responds to Image Presence in Ways Risk Cannot Explain
链接: https://arxiv.org/abs/2609.26174
作者: Haoyu Zhang,Yi Feng,Shibo Zheng,Zhuoxi Wang,Xiao Luo,Haowen Xu,Xiangchen Guan,Mohammad Zandsalimy,Shanu Sushmita
类目: Cryptography and Security (cs.CR); Artificial Intelligence (cs.AI)
备注:
Abstract:Vision-Language Model (VLM) safety is expected to depend on what a request asks for. We show that safety-aligned VLMs also key refusal on a property of a request’s form: whether an image is attached, holding everything the request asks fixed. Attaching a blank canvas - unreadable, unrelated to the request, identical across prompts - shifts refusal by tens of percentage points, with no defense in the loop. The shift is not blanket caution. Neutral instructions are almost unaffected while borderline-benign prompts move sharply, so the cost falls on sensitivity-adjacent traffic: benign questions about privacy, self-harm and violence. Attachment alone is sufficient, while the image’s properties set the price: a black canvas costs substantially more than a white one of identical size, and on an open checkpoint the carrying axis is pixel count. Nor is the shift under instructional control - telling the model the image is a placeholder to be disregarded removes only a fraction of it, and on one model asserting that an attachment exists moves refusal substantially with nothing attached. Attachment may correlate with risk in deployment; what these models do with it does not track risk. It is not the serving stack, since the same weights reached two ways behave alike, nor a property of VLMs as such, since several open-weight checkpoints show nothing. It belongs to particular aligned checkpoints, one of them open. It is also decoupled from what it buys: the canvas does prevent some attack success on a matched harmful set, but far less than it costs, and its sign is not fixed - on one open model the identical canvas makes the model markedly easier to attack. Image presence is not a default a deployer chose or priced; it is an uncontrolled variable inherited with the weights. Subjects: Cryptography and Security (cs.CR); Artificial Intelligence (cs.AI) Cite as: arXiv:2609.26174 [cs.CR] (or arXiv:2609.26174v1 [cs.CR] for this version) https://doi.org/10.48550/arXiv.2609.26174 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
[AI-49] he Free-Recipe Limit: Every Recipe Effect Measures Which Premise of an Idealised Learner Broke
链接: https://arxiv.org/abs/2609.26160
作者: Wenhui Chen,Jianlin Chen,Ziyao Lin,Chi Man Vong
类目: Artificial Intelligence (cs.AI)
备注: 42 pages, 9 figures, 17 tables (incl. appendices); code, pre-registration files, and result base to be released with the paper
Abstract:Fix a corpus and send recipe search to infinity: try every order of the skills, every arrangement from blocked to interleaved, every composition, and keep the best. Two quantities decide what that search was worth: the diameter of the reachable set it explores, and the resolution at which anyone can tell two endpoints apart. Where the diameter falls below the resolution, no amount of search converts into a decision, and the signature is not an absence of winners but winners that do not survive re-running. We measure this recipe-search wall with 761 fine-tuning runs on 12 base models (0.5B-14B, three pretraining families) over competition-mathematics skills: base checkpoints, supervised fine-tuning under AdamW, exact-match scoring at k=4. Within one coherent domain at fixed volume the three classical freedoms average 0.010-0.021 against a 0.019 floor, and the largest contrast, 0.0619, clears a three-seed resolution and then reads +0.010 and -0.015 on two reruns. The departure with a systematic answer is coherence: halving one pooled corpus and letting the halves write answers under incompatible but equally correct conventions moves arrangement from capability to allocation between conventions, by two orders of magnitude over a same-convention control, and writing the convention into the input switches the phenomenon off. The switch replicates on a second pretraining family and survives an independent re-execution of its own protocol, with a re-execution spread (0.087) smaller than the resolution a search-selected order cell carries (0.144). Order itself is a transient whose sign crosses zero three times inside a single run. Volume, the one lever nobody calls a recipe, is the one that reliably pays. A public scorecard grades all 26 pre-registered claims: 18 supported, 5 failed, 2 untested, 1 mixed.
[AI-50] oward User-Mediated Self-Repair in Ubiquitous Robots Through Goal-Oriented Agent ic AI
链接: https://arxiv.org/abs/2609.26157
作者: Morten Roed Frederiksen
类目: Artificial Intelligence (cs.AI); Robotics (cs.RO)
备注: Preprint submitted version, 8 pages
Abstract:Ubiquitous robotic systems often lack traditional visual interfaces, necessitating resilient natural language interaction for maintenance and repair tasks. This paper presents a goal oriented agentic AI architecture designed to enable non-expert users to perform technical repairs through situated dialogue. The framework utilizes a multi-layered approach that decouples high-level strategic planning from reactive conversational execution to transform unconstrained human instructions into a structured hierarchy of goals. We conducted a study involving twenty participants to evaluate the system’s efficacy using a physical hardware testbed. The architecture achieved a 95% task completion rate, and participants reported positive self-efficacy following real-time guidance that adapted to conversational diversions and linguistic variations. A comparative analysis with an online baseline revealed that the transition to a physical environment significantly decreased perceived social presence (p=.0005), and trust and competence, (p=.037), while the agentic framework remained robust throughout the interaction. These findings indicate that goal oriented agentic AI can support the sustainability of body-worn technologies by empowering users to perform critical maintenance in ubiquitous contexts.
[AI-51] Unanimity Without Persuasion: A Single Round of Debate Erases the Disagreement That Verification Needs
链接: https://arxiv.org/abs/2609.26145
作者: Yang Shu
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:A debate panel can become unanimous without becoming more correct. This is dangerous for downstream safeguards: a substituted verification ballot can change only narrow-margin votes, while richer arbiters lose disagreement as a natural targeting signal. We show that one debate round can erase that resource without requiring persuasion. Tracking a heterogeneous 7-judge panel through a blind round and three debate rounds on 600 code-correctness candidates, unanimity on a fixed cohort jumps from 39.5% to 95.2% in round 1 (93.1% of the total collapse), while accuracy moves by less than one point and 96.3% of verdict flips follow the displayed peer majority. An execution-based verification ballot corrects 8 of 2,037 pre-debate candidate-substitution instances but changes zero in every later round; by round 3 every wrong decision is unanimous, erasing dissent that had flagged two-thirds of the panel’s errors. Identical-cohort controls explain why: no-peer reconsideration reproduces 79.6% of the collapse, real labels without reasoning reproduce 91.5%, and random labels steer flips toward whatever they display; the full-debate condition adds 4.3 percentage points over labels only (clustered 95% CI 0.7–8.1). The one-round collapse reproduces in two additional real runs and two fake-label seeds, remains under panel sizes 3–7, and appears in MATH-500. Parse failures concentrate on contested candidates ( p0.001 ), making attrition non-ignorable. The design implication is operational: verify before any second-pass evaluation or peer exposure, and never treat post-debate unanimity as independent evidence of reliability.
[AI-52] When Verifiers Vote Backwards under Verdict Substitution: Signed Pivotal Value in Correlated Self-Consistency
链接: https://arxiv.org/abs/2609.26144
作者: Yang Shu
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Replacing one ballot can change a majority decision only on queries decided by a single vote; this structural fact requires no independence assumption. We study the sign of that change using a labeled, verdict-style intervention: one correctness signal replaces one correctness-indicator ballot in k=7 self-consistency panels. This diagnostic intervention is not identical to deployed answer-identity plurality. A primary MATH-500 experiment ( n=570 ) gives a different-model verifier a +24.2 pp pivotal gain, whereas a role-reversed configuration gives -11.2 pp; an exploratory code stress test (14 pivotal rows across 9 tasks) gives -24.5 pp. An exact signed-gain decomposition accounts for all observed signs through the verifier’s state-specific accuracy and the composition of the two one-vote tally states, rather than global accuracy or model provenance. Same-source signals lose accuracy on the pivotal stratum (65 \to 44% in the primary configuration), while error correlations provide a descriptive error-association diagnostic. Controlled degradation and a k\in\3,5,7\ subset sensitivity analysis probe the stability of the observed pattern around this accounting. Under the evaluated ties-incorrect answer-identity plurality analysis, the structural zero and strong-verifier benefit persist, but the role-reversed harm attenuates to -0.9 pp and is not significant. The results therefore establish harmful verdict substitution, not universally harmful deployed plurality, and motivate a testable but unverified hypothesis for negative process-reward-model weights.
[AI-53] StepTrigger: Contact-State-Triggered Backdoor Attacks on VLM-Powered Legged Robots
链接: https://arxiv.org/abs/2609.26131
作者: Jiageng Zhang,Doniyorkhon Obidov,Kaichen Yang
类目: Robotics (cs.RO); Artificial Intelligence (cs.AI); Cryptography and Security (cs.CR)
备注:
Abstract:Large language models and vision-language models are increasingly used as high-level planners in robotic systems, using task goals and sensor summaries to select navigation or manipulation actions. This creates a new backdoor surface: a compromised planner can behave normally in most runs, yet change its target selection when a hidden trigger is present. Prior attacks on LLM-powered or embodied agents mainly rely on triggers that appear in language, camera-visible objects, scene semantics, or specific sequences of past actions. This paper presents StepTrigger, a contact-state-triggered backdoor attack for VLM-powered legged robots. The trigger is not a prompt token or a visible marker. It is produced by pressure and foot-ground contact patterns that arise when a Unitree Go1 quadruped walks across a dense terrain patch. Unlike conventional visual or textual triggers, contact signals are inherently noisy and may also arise during benign locomotion. To avoid treating every pressure anomaly as a trigger, StepTrigger learns a selective backdoor policy from multimodal robot state, using incidental pressure events as benign examples and dense-patch contacts as poisoned examples. In a stratified offline evaluation, the trained planner achieved 98.75% clean behavior preservation, 92.50% false-trigger rejection, 76.25% true-trigger activation, and 89.17% overall parsed behavior accuracy. These results reveal a backdoor surface in proprioceptive and contact channels that is not captured by defenses focused only on language, vision, or action history. Subjects: Robotics (cs.RO); Artificial Intelligence (cs.AI); Cryptography and Security (cs.CR) Cite as: arXiv:2609.26131 [cs.RO] (or arXiv:2609.26131v1 [cs.RO] for this version) https://doi.org/10.48550/arXiv.2609.26131 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
[AI-54] he Cost of Conservation: Coordination-Memory Laws for Exact-Support Generation
链接: https://arxiv.org/abs/2609.26126
作者: Zhen Zhang,Amr Alanwar
类目: Artificial Intelligence (cs.AI); Computational Complexity (cs.CC)
备注:
Abstract:Many AI systems make decisions locally, even when every realized output must obey an additive conservation law, such as selecting exactly a fixed number of items. This constraint can be statistically invisible: small subsets of a balanced fixed-budget output look increasingly independent, yet communication-free coordinate-parallel generation needs exponentially many pre-shared plans, while a sequential exact sampler needs only logarithmic memory. We study product measures conditioned on additive conservation laws in the intermediate regime where one plan is selected before a fixed-order pass, every plan is a bounded-state stochastic executor whose support is entirely legal, and the mixture of plan laws approximates the target distribution in total variation. Our main result identifies the optimal asymptotic selector rate, up to constant factors, with the killed spectral profile of the conservation-difference walk. The resulting coordination cost decreases as an inverse power of live-state width, with an exponent determined by intrinsic conservation rank rather than alphabet size; the law extends to noncentral budgets and heterogeneous local scores. The converse is driven by a state-versus-resource-sum obstruction, while a rate-matching construction compiles discrepancy control into exact-support finite-state plans. Complementary results characterize block-parallel plan complexity and the benefit of programmable output order. Together, these results show precisely how online memory substitutes for front-loaded coordination in exact-support generation.
[AI-55] When Big Data Becomes a Curse: Spatial Heterogeneity and the Limits of Learning from Passive Acoustic Monitoring Data
链接: https://arxiv.org/abs/2609.26125
作者: Gabriel Spadon,Wayne Renaud,Priyanka Aravindan
类目: Artificial Intelligence (cs.AI); Sound (cs.SD)
备注:
Abstract:Passive Acoustic Monitoring produces large archives whose recordings are clustered by deployment, season, station identifier, and acquisition configuration. We analyze 908,072 AIS-labeled 679-second recordings from 38 deployments, 20 Atlantic Canadian station identifiers, and 21 receiver positions. The AIS-contact prior varies by more than 200-fold, and per-deployment screening distributions require local interpretation. Bidirectional cross-season transfer over the 18 station identifiers observed in both seasons predicts station identity above the 5.56% uniform-chance level, with balanced accuracy of 15.9% for AIS-contact and 16.4% for no-AIS-contact recordings. The same descriptors predict the two hydrophone models at 74.1% and 85.6% balanced accuracy, respectively, but hydrophone model is strongly confounded with season and other deployment-level acquisition differences. On a retrospectively screened and capped benchmark of 54m507 recordings, repeated station-grouped holdout yields an ROC-AUC of 0.612 with a station-bootstrap 95% interval of 0.582 to 0.648, compared with 0.661 under a random-window diagnostic. Their paired difference is 0.049 (0.040 to 0.056). Removing raw energy changes unseen-station ROC-AUC from 0.612 to 0.603, while an exploratory training-station scale analysis is non-monotonic. These results show that random-window validation overstates transfer to unseen station identifiers in this corpus. They support dependence-aware validation and broader independent spatial sampling, while spatial-expert models remain a hypothesis rather than an established remedy.
[AI-56] MAC-RRG: Iterative Multi-Agent Collaboration for X-ray Radiology Report Generation
链接: https://arxiv.org/abs/2609.26124
作者: Futian Wang,Yuhan Qiao,Xiao Wang,Dan Xu,Yuehang Li,Zhixiang Guo,Yaowei Wang,Jin Tang
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Despite the remarkable progress of LLM-based and knowledge graph-augmented Radiology Report Generation (RRG) methods, existing techniques still suffer from inherent defects. Conventional LLM-only models lack structured medical prior knowledge, resulting in frequent medical hallucinations and low diagnostic interpretability. Current knowledge graph-enhanced schemes adopt static one-round knowledge fusion with single-source knowledge, incapable of dynamic knowledge updating according to generation feedback. This paper proposes a novel Multi-Agent Collaborative iterative framework for X-ray Radiology Report Generation, termed MAC-RRG. Inspired by multi-agent technology, our framework constructs a closed-loop optimization paradigm based on task decoupling and collaborative reasoning. Specifically, the framework first generates a preliminary radiology report from input X-ray images via a vision encoder and a basic LLM. Subsequently, a multimodal knowledge graph (MM-KG) agent mines structured disease correlation and anatomical knowledge from medical knowledge graphs, while an auxiliary knowledge agent extracts unstructured domain knowledge from public medical databases. The multi-source knowledge acquired by dual agents is fused and embedded to guide the LLM in iteratively refining the initial report. Extensive quantitative and qualitative experiments on mainstream X-ray RRG datasets, including IU X-ray, MIMIC, and CheXpert Plus, fully verify the superiority of our proposed method. The source code and pre-trained models have been released on this https URL
[AI-57] FairMon: A Tool for Monitoring and Visualizing Algorithmic Fairness
链接: https://arxiv.org/abs/2609.26123
作者: Jan Baumeister,Bernd Finkbeiner,Vladimir Krsmanovic,Frederik Scheerer,Julian Siber,Tobias Wagenpfeil
类目: Artificial Intelligence (cs.AI); Logic in Computer Science (cs.LO)
备注: Accepted at RV 2026
Abstract:Runtime monitoring has recently been proposed as a rigorous method for analyzing algorithmic fairness of autonomous decision systems used in critical scenarios such as credit lending, job application, and the criminal justice system. Prior work has shown that runtime monitoring, in principle, can be an effective technique for establishing the kind of human oversight required by legislation such as the EU Artificial Intelligence Act. In practice, the available monitoring tools have not been developed with this application in mind and display several critical shortcomings in these scenarios. In this paper, we present FairMon, a runtime monitoring tool tailored to fairness analysis of high-stakes decision systems. FairMon uses RTLola as a flexible specification language for monitors, which we have extended with conditional probability operators that allow for concise descriptions of algorithmic fairness properties. The tool also features a real-time visualization of intermediary values, enabling human insight into the dynamics of the monitored system.
[AI-58] Certified Mechanistic Interpretability: Lifting Single-Input Findings to Bounded Neighbourhoods
链接: https://arxiv.org/abs/2609.26112
作者: Zhen Zhang,Yanliang Huang,Peng Xie,Wenyuan Wu,Amr Alanwar
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
备注:
Abstract:Mechanistic interpretability reverse-engineers transformer circuits one input at a time, leaving observed mechanisms without guarantees over bounded input neighbourhoods. We address this gap with a framework based on constrained polynomial-zonotope (CPZ) propagation that lifts mechanistic-interpretability observations from a single input to certified statements over a bounded set of perturbations. Three internal-attention queries (top- k stability, evidence mass, and attention entropy) are formulated as tractable programs over the simplex of attention weights, and CPZ propagation through transformer blocks is shown to preserve the softmax simplex and the LayerNorm zero-mean identity exactly. A recursive Jacobian zonotope construction extends the same certificates across layer depth by linearising the block stack at the input and avoids per-layer generator growth. We instantiate the framework on transformer attention; the resulting certificates offer a way to sharpen mechanistic statements that single-input inspection cannot resolve on its own, and to inform downstream decisions in regimes where empirical heuristics may be misleading.
[AI-59] Early Prediction of Pathological Complete Response to Neoadjuvant Chemotherapy Using Temporal Deep Learning on DWI
链接: https://arxiv.org/abs/2609.26106
作者: Pablo García Marcos,Md. Tarequl Islam,Paula Puerta González,Guillermo Lorenzo,Héctor Gómez,Covadonga del Camino,Angel Rio-Alvarez,Víctor M. González
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Early identification of non-responders to neoadjuvant chemotherapy (NACT) is crucial for timely treatment adaptation in breast cancer. However, many existing predictive models rely on multiparametric magnetic resonance imaging (MRI), late treatment time points, or extensive clinical data, which limits their applicability. This study proposes a deep learning framework for early prediction of pathological complete response (pCR) using only diffusion-weighted MRI (DW-MRI) acquired at baseline and after the first NACT cycle. This framework feeds cropped tumor-centered patches to an EfficientNet-based temporal model that directly learns tumor shape and local tissue characteristics without explicit radiomic feature engineering. The model, trained with 10-fold cross-validation, achieved an area under the receiver operating characteristic curve (AUC) of 0.90 for pCR prediction after one cycle, providing actionable information after a single treatment cycle while avoiding gadolinium administration and reducing dependence on heterogeneous clinical data. By focusing on the baseline-to-first-cycle window instead of later stages, the approach supports earlier escalation or de-escalation of NACT, and its exclusive reliance on DW-MRI facilitates protocol standardization, multi-centre deployment and privacy-preserving data sharing. These results demonstrate that DW-MRI-based deep learning on tumor-centered patches constitutes a minimally invasive, clinically deployable strategy for early pCR prediction, with direct implications for personalized treatment adaptation in neoadjuvant breast cancer therapy.
[AI-60] he Architect the Adversary and the Judge: Closed-Loop Generation of Standards-Aligned Assessment Items at Scale
链接: https://arxiv.org/abs/2609.26087
作者: Wenhui Chen,Ziyao Lin,Jianlin Chen,Peiji Long,Chi Man Vong
类目: Artificial Intelligence (cs.AI)
备注: 31 pages, 8 figures, 16 tables. Companion paper: arXiv:2608.01000
Abstract:We present CLAIM, a production pipeline for K-12 assessment-item generation coupling a two-stage generate-then-attack protocol (the model drafts as a “curriculum architect”, then re-enters the same conversation as a hostile adversarial reviewer), bi-directional few-shot conditioning on accepted and rejected items, the latter carrying the evaluator’s diagnosis, and a knowledge dictionary of 44,844 error-correction rules mined from that feedback and retrieved per standard and item type. Across 43,227 scored items over 755 Common Core ELA standards, three item types, and ten LLMs, the pipeline reaches a 97.8% expert-evaluator pass rate on a 9,074-item production run. We then ask what that rate certifies. Re-scoring a stratified sample with three judges from other vendors, blind to the deployed verdict, reproduces the format ordering under every judge and recovers a larger open-set deficit than the deployed evaluator does; but agreement on the accept/reject binary is weak at production prevalence (kappa about 0.13), and the judges agree with each other no better. The level is therefore judge-relative, and with no student-response data our quality evidence is evaluator-judged throughout. The corpus also exposes a robust asymmetry. Multiple-choice and multiple-select generation saturate at 98% or above for both frontier models under a dozen static rules, whereas fill-in-the-blank generation is capability-tiered (82.8-96.7% across five models under a matched rule set, standards, and judge) and plateaus under prompt-only optimization, with error mass shifting between answer-key over-inclusion and omission as rules accumulate. We analyze this as open-set boundary determination, a task autoregressive decoders are structurally ill-equipped to solve, and show the asymmetry recurring when the evaluator itself is distilled: fail-recall rises from 8% to 63% while F1 saturates at 0.25.
[AI-61] Selection-Invariant Communication Compilers for Privacy-Aware Multi-Agent LLM Workflows
链接: https://arxiv.org/abs/2609.26076
作者: Jinghan Xu,Longze Fan,Zeyuan Wang,Xinjin Li,Hankai Liu
类目: Artificial Intelligence (cs.AI); Cryptography and Security (cs.CR)
备注:
Abstract:Structured multi-agent workflows exchange intermediate messages whose content and form can reveal private state even when the final output is safe. We identify selection-channel leakage: after authorization fixes what may be released, a private-state-aware choice among semantically valid realizations creates an additional inference channel. We introduce the selection-invariant communication compiler(SICC), which constrains this post-authorization representation kernel rather than prescribing templates. Any deterministic or independently public-randomized generator satisfying the invariant is valid; requirement-indexed canonical forms are one auditable implementation. We prove a compositional communication-layer guarantee: authorization, public-only form generation, and a dependency-safe utility gate make the emitted transcript reveal no information beyond the complete authorized view. Private-state-aware selection remains vulnerable after surface-disjoint and length-matched controls. Across 132 AgentLeak communication replays and 100 executable LangGraph tasks, deterministic SICC retains complete protocol utility without a positive excess-gain signal; independent public randomization preserves the same result in AgentLeak and 480 controlled cases.
[AI-62] Policy-Backed Selective Regeneration under Tainted Inter-Agent Communication
链接: https://arxiv.org/abs/2609.26072
作者: Jinghan Xu,Longze Fan,Zeyuan Wang,Xinjin Li,Hankai Liu
类目: Cryptography and Security (cs.CR); Artificial Intelligence (cs.AI)
备注:
Abstract:Inter-agent communication is essential to multi-agent language-model systems, yet a single message may combine task-critical information with instructions not authorized by the original request. Prompt-based defenses leave enforcement to models exposed to adversarial messages, while indiscriminate message removal discards useful information. We introduce Executable Semantic Commitments with Clean-Room Recovery (ESC-CR), a policy-backed framework for secure inter-agent code generation and recovery. It separates message claims from authorization, constructs executable commitments from trusted tasks, evidence, and policy, and enforces them at an external release boundary. Upon a violation, ESC-CR taints the responsible message and rejected artifact, reconstructs a clean context from evidence-backed task information, and regenerates under the same policy. We evaluate ESC-CR across communication-essential and standard code-generation benchmarks, multiple model families and communication topologies, and adaptive attacks spanning direct, obfuscated, and verifier-aware payloads. Results show that polluted-context retry frequently fails to remove unauthorized influence, while complete message removal can discard information required by communication-essential tasks. ESC-CR preserves evidence-backed claims while suppressing unauthorized releases under matched computational budgets, and the same design transfers to end-to-end agent trajectories.
[AI-63] RankCert: When Can Simulated Learners Safely Select an AI Tutor? Robust Decision Certification Under Structural Uncertainty
链接: https://arxiv.org/abs/2609.26069
作者: Nizam Kadir
类目: Artificial Intelligence (cs.AI); Computers and Society (cs.CY); Machine Learning (cs.LG)
备注: 19 pages, 5 figures, 7 tables. Submitted to ACM Transactions on Intelligent Systems and Technology (TIST)
Abstract:Simulation-based tutor selection can be unstable when predictively adequate learner models imply different policy rankings. RankCert certifies one of eight equal-budget tutoring policies only when model-averaged utility, probability-best, posterior regret, cross-domain rank, family coverage, and leave-one-domain-out and leave-one-visible-family-out averages support the same candidate; otherwise it abstains. We evaluated RankCert in 1,280 frozen held-out settings spanning five rotating held-out oracle families, 64 scenarios per family, and four cohort sizes. Calibration used a licensed, de-identified EdNet-KT1 derivative with 5,000 learners and 590,056 retained responses; all five family representatives passed the frozen adequacy gate. Minimum-domain mean pairwise top-1 agreement was 0.272917 (95% CI [0.253646, 0.293229]), showing substantial structural disagreement. Cohort-noise variance decreased from n = 30 to n = 300, while the structural family share remained nonzero. RankCert reduced total held-out decision loss relative to full-coverage point selection by 0.006605 normalized-outcome units (95% CI [0.004859, 0.008407]). At comparable coverage, however, it did not reduce selective risk relative to a confidence-gated point certificate (difference -0.000213; 95% CI [-0.003238, 0.002384]; Holm p = 0.929654). Certification occurred in 3.75% of settings and only in stable scenarios; RankCert abstained in every ambiguous, misspecified, and structural-conflict setting. “Safe” denotes only benchmark-scoped decision certification under the declared utility and uncertainty set; no human-learning, causal, deployment-effectiveness, or general-safety claim is made.
[AI-64] ChainUQ: Reasoning Consistency-Aware Uncertainty Quantification for Large Language Models
链接: https://arxiv.org/abs/2609.26060
作者: Dahai Yu,Rongchao Xu,Lin Jiang,Ximiao Li,Guang Wang
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:While large language models (LLMs) exhibit impressive reasoning capabilities, response-level confidence may remain unreliable when intermediate claims conflict with the final conclusion. Therefore, effective uncertainty quantification (UQ) is required to capture logical inconsistencies within the reasoning chain, not just the correctness of the final output. Current approaches have two major limitations: (1) their reliance on token-level probabilities fails to capture reasoning consistency, and (2) they lack mechanisms to dynamically calibrate confidence using the structural logic of the generated chain. To advance existing research, we introduce ChainUQ, a reasoning consistency-aware uncertainty quantification framework for LLMs. ChainUQ consists of two key technical components: an alignment-aware lightweight UQ module that estimates a raw intrinsic model confidence score from frozen features aligned to the final conclusion, and a reasoning consistency-aware calibrator that refines this score using reasoning-chain consistency evidence. Evaluations across diverse in-distribution and out-of-distribution benchmarks show that ChainUQ consistently improves response-level uncertainty estimation, achieving an average 3.1% relative gain in AUROC and up to 45.0% relative reduction in ECE, and can be directly transferred to new settings without additional fine-tuning.
[AI-65] Observing the Conduct of Systematic Reviews with Generative AI Support: An Experience Report from a Graduate Software Engineering Course
链接: https://arxiv.org/abs/2609.26057
作者: Danilo Monteiro Ribeiro,Gilberto Sussumu Hida
类目: Computers and Society (cs.CY); Artificial Intelligence (cs.AI); Software Engineering (cs.SE)
备注: Accept in SBES Education
Abstract:Context: Secondary studies are fundamental practices in Evidence- Based Software Engineering, but teaching them requires activities that expose students to authentic methodological decisions. Objective: This paper reports an experience in a graduate course in which ten doctoral students in Software Engineering, organized into three groups, piloted secondary studies with and without support from generative AI. Method: A single-day classroom session was organized and observed, in which the groups conducted pilot systematic reviews with and without generative AI support. Classroom observations, produced artifacts, and interaction threads with assistants configured in ChatGPT were analyzed to reconstruct how each group appropriated the technology throughout the activity. Results: LLMs reduced initial barriers, accelerated the generation of alternatives, and made methodological problems more explicit, but they also favored excessive delegation, superficial validation, operational difficulties, and a shift in focus from conducting the SLR to using the tool. Conclusion: The experience offers a situated, observational account of how doctoral students engaged with generative AI during a systematic review activity, and the resulting insights also inform the design of a subsequent controlled study. The findings indicate that generative AI can support practical learning about SLRs, provided that its use is accompanied by human supervision, decision records, and critical reflection on its limitations.
[AI-66] Canonical locks that encode part-whole hierarchies
链接: https://arxiv.org/abs/2609.26046
作者: Rajat Modi,Yogesh Singh Rawat
类目: Artificial Intelligence (cs.AI)
备注: Work in Progress
Abstract:One of the challenges in representational learning is how to encode part-whole hierarchies in a neural net. Prior works rely on flattening tree-like structures into string-like sequences and training a sequence-to-sequence model via autoregression. While such a representation works for parse-trees in NLP, it is not entirely clear how to make it work for images. Thus, we propose a geometric primitive called canonical locks. The key idea is that parts/wholes can be modelled as higher-dimensional vectors ( d \geq 4 ), and information can be encoded in their relative phase differences. Inductively, the net consists of positionally-bound bottom-up and top-down neural fields, which drive each other to achieve a state of thermal equilibrium. Additionally, we show the existence of a few symmetrical configurations in the net. The computational iterations taken to break these symmetries depend on the angle between parts/wholes arranged on a disk (or more precisely a ring) in higher dimensions. It also appears to have connections to the psychological phenomenon of mental rotation. Comments: Work in Progress Subjects: Artificial Intelligence (cs.AI) Cite as: arXiv:2609.26046 [cs.AI] (or arXiv:2609.26046v1 [cs.AI] for this version) https://doi.org/10.48550/arXiv.2609.26046 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
[AI-67] xWhyL: Causal Interactive Learning
链接: https://arxiv.org/abs/2609.26037
作者: Nicholas Tagliapietra,Florian Peter Busch,Moritz Willig,Matej Zečević,Lavdim Halilaj,Juergen Luettin,Kristian Kersting
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Machine Learning (stat.ML)
备注:
Abstract:Explanations are central to causal reasoning, and cognitive science has long established that the human drive to explain is itself a mechanism for learning about causality. Despite this, learning from those abductive signals is largely ignored in artificial intelligence. While explainable AI (XAI) increasingly draws on causal models to generate explanations, the converse direction about what explanations can do for causality remains largely unexplored. To fill this gap, we propose xWhyL, a formal framework connecting causality and XAI by learning causal models from explanations. We develop a mathematical theory that translates explanations into a learning signal complementary to observational data, and demonstrate how it enables overcoming the limits of observational causal discovery. As explanations can be derived from incorrect beliefs and clash with data, a tension we call the Causal Tug-of-War, we prove conditions under which our framework rejects misspecified explanations rather than absorbing them. Our practical instantiation, Causal Interactive Learning (CIL), shows how expert explanations can efficiently support causal discovery and distinguish correct from incorrect explanations.
[AI-68] CQ4OE: A benchmark for assessing LLM -assisted ontology generation from competency questions
链接: https://arxiv.org/abs/2609.26029
作者: Jiayi Li,Ziyuan Wang,Daniel Garijo,María Poveda-Villalón
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Ontology generation from Competency Questions (CQs) is a central yet labor-intensive phase of Ontology Engineering. While large language models (LLMs) offer promising automation capabilities, current evaluations remain fragmented. Task formulations are heterogeneous, gold standards often lack fine-grained CQ provenance, metrics conflate lexical overlap with structural and logical adequacy, and reference ontologies are not always explicitly designed around the evaluation CQs. Here, we address these limitations with CQ4OE, a benchmark for the systematic and reproducible evaluation of LLM-based ontology generation from CQs. For each ontology in the benchmark, we build a CQ-driven gold OWL ontology with explicit provenance linking each CQ to the classes, properties, and axioms required to answer it. From this resource, we define two complementary evaluation tasks. CQ2Term supports term-level evaluation of CQ-specific class and property prediction over 99 CQs, and CQ2Onto supports ontology-level evaluation over 118 CQs, including hierarchy, property modeling, and axiom-level structure. We demonstrate CQ4OE with experiments using nine LLMs under zero-shot, iterative, and multi-agent generation strategies, showing that LLMs recover explicit vocabulary terms more reliably than creating ontologies, particularly in property modeling, hierarchy construction, and axiom generation.
[AI-69] REVE: Efficient Hallucination Correction for Large Audio-Language Models via Reused Encoder States
链接: https://arxiv.org/abs/2609.26028
作者: Hongjin Song,Jiasheng Kuang,Xinyu Yang,Qiuyu Fang,Ziyu Wu,Guowu Tan,Xiang Xie
类目: ound (cs.SD); Artificial Intelligence (cs.AI)
备注:
Abstract:Large audio-language models may mention acoustic events that are absent from the input. A separate audio event detector can verify these mentions, but doing so requires a second audio encoder and a separate forward pass. We propose Reused Encoder States for Verifying Events (REVE), a lightweight method that uses states already computed by the target model. One readout summarizes class scores across audio frames, while another uses pooled states from four consecutive frame intervals. Class-aware score fusion combines their outputs to verify generated event mentions without encoding the audio again. On AudioSet, REVE removes 92.9% of label-unsupported mentions under a faithful-mention recall constraint. With fewer added parameters and no second audio-encoding pass, REVE achieves a reduction comparable to those of CED-Tiny and CED-Base. Its complete verification latency is about 1/18 of the CED-Base path. Results on controlled DESED mixtures and different target-model architectures further confirm the effectiveness of encoder-state reuse.
[AI-70] Reciprocal Collaboration: how lessons from convergence in GLAMs can enhance interdisciplinary AI research
链接: https://arxiv.org/abs/2609.26023
作者: Amber L. Cushing,Suzanne Little,Giulia Osti
类目: Computers and Society (cs.CY); Artificial Intelligence (cs.AI)
备注: 19 pages, 1 figure
Abstract:The need for collaboration between diverse fields of research is increasingly recognised as important by research funding agencies. A significant driver of this need is the current revolution in artificial intelligence (AI) and related technologies. There is a growing interest in the potential impact of AI in different fields including the methodologies they use and the resulting advances in new knowledge, new access and enhanced productivity. However, there is also a corresponding increase in concern about the fundamentals of AI technologies and the way in which trans and/or interdisciplinary research is approached. The resulting collaboration too often ends up as a one-way street where the domain partner acts only as an information provider. For example, the contribution of the AHSS partner might be limited to providing insight about ethics and/or the technology partner may only provide a service to build applied AI-based solutions. In response to this problem, we propose a reciprocal approach to collaboration where both partners seek to understand, cooperate and identify jointly significant impacts. In this paper we explore this relationship between cultural heritage institutions (GLAMs), Arts, Humanities Social Sciences (AHSS) research and technology-led AI research, especially the impact of current technological advances in AI. Drawing from the history of convergence in GLAM studies, we propose five key practices to form a framework for greater understanding across this divide.
[AI-71] Compiling Sufficient Governance Context from Declared Losses and Reachable States: Exact Observation-Contract Synthesis with Cardinality and Cost Objectives
链接: https://arxiv.org/abs/2609.26016
作者: Gaston Besanson
类目: oftware Engineering (cs.SE); Artificial Intelligence (cs.AI); Logic in Computer Science (cs.LO)
备注: Code, data, preregistration tags, review record, and independent reproduction (repository issue #3): this https URL . Artifact DOI: https://doi.org/10.5281/zenodo.22884173
Abstract:We call the object this paper derives and certifies a minimal sufficient governance context: given a finite reachable-state model, a deterministic declared verdict, and candidate observable attributes, we compute sufficient observation sets, distinguish attributes that are individually indispensable from contracts that are jointly sufficient, and select among sufficient contracts under a cardinality or declared-cost objective. An observation contract is a set of candidate attributes whose values determine the declared verdict on every reachable state; an authority contract is one selected under an objective and bound to a gate schema. We synthesize every inclusion-minimal sufficient contract where exhaustive enumeration is affordable, and a minimum-cardinality or minimum-cost contract by SAT/MaxSAT encoding otherwise, checking sufficiency directly. On a constructed code/cloud domain, the individually-indispensable core is not sufficient as an observation contract and two distinct reducts exist; a preregistered cost model separates them exactly. On a second, larger, constructed domain, the same pattern recurs, but that domain’s cost model does not separate the alternatives: a fully explained cost tie, reported as found. We measure discernibility-family scaling where exhaustive enumeration is confirmed infeasible within a registered timeout, while SAT/MaxSAT synthesis solves in well under a second; MaxSAT showed no measured cardinality advantage over plain SAT. AuthorityBench compares four baselines across three domains; the declared-only baseline is not exactly sufficient on any. Every selected contract is checked for sufficiency, with a counterexample on failure and a check summary, not a portable certificate, on success – the compiler-focused scope of a two-scope table; an independently specified end-to-end case study is registered follow-up work, not claimed here.
[AI-72] VideoX-Qwen : Data-Centric Instruction-Based Video Editing
链接: https://arxiv.org/abs/2609.26015
作者: JJiahang Li,Dingbao Shao,Xinyu Chen,Song Wu,Jiang Lin,Duo Li,Yuhang Liu,Jiaxin Hu,Shengrong Gu,Ying Tai,Zili Yi
类目: Artificial Intelligence (cs.AI); Graphics (cs.GR)
备注: Technical report
Abstract:Progress in general-purpose video editing depends on constructing large-scale paired supervision and effectively adapting video-generation backbones to instruction-driven editing. Unlike video generation, video editing must execute a requested transformation while preserving unrelated subjects, scene structure, motion, and temporal continuity. We present VideoX-Qwen, an integrated data-construction and model-training framework for general instruction-based video editing. Our scalable production pipeline organizes specialized generation and understanding models into complementary routes for addition, removal, replacement, and attribute editing, followed by quality screening and instruction enrichment. It produces more than 1.2 million directional video-editing records, including over 400,000 records in each major task group, with an automatic acceptance rate of 89%. The resulting corpus provides broad and structured coverage of common editing operations through a unified source-instruction-target interface. We further develop a unified Qwen-Wan editor that combines multimodal semantic conditioning with dense source-video latent guidance. A progressive image-video training strategy aligns the multimodal instruction interface, adapts the video generator to source-conditioned editing, and refines output quality with selected high-resolution data. In a 100-example comparison with UniVideo and Kling O1, VideoX-Qwen achieves the best mean result on nine of eleven reported metrics, including instruction following, editing quality, content preservation, structural and perceptual similarity, and video-distribution quality. Together, the large-scale data-production system and unified training framework provide a practical foundation for more capable instruction-driven video editing.
[AI-73] Skytopia: Monocular Drone Navigation with Action-Conditioned Latent World Models
链接: https://arxiv.org/abs/2609.26007
作者: Yuhang Zhang,Rangya Zhang,Yujing Shang,Zhuoyuan Yu,Weiying Wang,Steven Yang,Qingsong Yan,Chao Yan,Mir Feroskhan
类目: Robotics (cs.RO); Artificial Intelligence (cs.AI)
备注:
Abstract:Monocular drone navigation requires reaching a goal in an unseen environment from a single forward-facing camera, which offers few cues for depth and scale. World models address this by modelling how observations evolve under actions, but they are built to be executed: the prediction is produced at deployment and fed back into action generation at every control step. We argue that what a policy needs from a world model is not the prediction but the representation required to produce it: in flight the executed action explains almost all of the change between observations, so prediction reduces to reprojecting a static scene under a known displacement. We therefore introduce skytopia, a policy built on an action-conditioned latent world model, and the 3D Gaussian Splatting platform on which it is trained. A forward objective predicts the representation of the next observation from the intended motion, and an inverse objective recovers that motion from the predicted transition. Because the prediction never reaches action generation, the predictor is discarded and one policy serves point-goal, image-goal, and goal-free navigation. Simulation experiments show that skytopia outperforms every baseline under all three specifications, attaining 57.8%, 66.0%, and 49.0% success rate, while discarding the predictor removes 59.4% of the inference cost. The same policy is subsequently deployed on a physical drone without fine-tuning and reaches goals in indoor, open outdoor, and woodland environments.
[AI-74] Interweaving Marginals into Multivariate Sample Paths: Training-Free Dependence Construction for Probabilistic Time Series Foundation Models
链接: https://arxiv.org/abs/2609.25980
作者: Jinmyeong Choi,Jinkwan Jang,Seul Lee,Taesup Kim
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
备注:
Abstract:Probabilistic time series foundation models (TSFMs) provide coordinate-wise predictive distributions, but these marginals do not determine a joint distribution over multivariate future trajectories. We study training-free coupling of frozen TSFM marginals into multivariate forecast sample paths. Our primary evaluation fixes the empirical marginal sample multiset at every channel–horizon coordinate across methods, isolating the effect of coupling alone. Historical temporal and channel relations substantially improve their corresponding dependence diagnostics. The same pattern persists when the fixed-marginal constraint is removed and paths are sampled directly, and remains present under native multivariate backbone inference. These results support treating dependence reconstruction as a distinct post-processing problem for probabilistic TSFMs.
[AI-75] CausalLoss-Fin: Attributing Financial-Agent Loss to Decisions and Infrastructure Faults
链接: https://arxiv.org/abs/2609.25960
作者: Abhishek Sharma
类目: Artificial Intelligence (cs.AI)
备注: 8 pages, 3 figures, 5 tables. Code and reproducibility materials: this https URL
Abstract:When an agent handling a payment exception loses money, the agent-step attribution methods this paper compares against will name one of its actions. They will do so even when a settlement message was dropped and the agent never had a chance: they intervene on agent actions and do not expose infrastructure faults as intervenable variables, so every dollar they explain is charged to a decision. We take a benchmark whose fault process is explicit and replayable, decompose each episode’s realised delivery schedule into named, individually repairable messages, and intervene on both the agent’s choices and the infrastructure’s. A telescoping identity splits any policy’s loss exactly three ways: an infrastructure effect, a policy differential against the best implementable policy, and a reference-policy residual. Two of the three can be negative, so none is a share; Shapley then divides the first into signed allocations over individual messages. One result is structural and needs no corpus: an agent-only baseline identifies no infrastructure cause, because its model contains no variable that could name one. What 545 planted episodes across 3 policies measure is the size of that consequence. It misfiles 100% of infrastructure episodes and charges 114,383.40 to the agent. Repairing what it names recovers 0.0% of the available loss; repairing a minimal sufficient set recovers 100.0%. Scoring messages one at a time is not merely imprecise: 27.8% (95% CI: 23.3–32.3%) of episodes do not decompose additively. We evaluate deterministic programmatic policies rather than language-model agents, which is what makes replay exact and which limits external validity to stochastic agents. The prevalence figures are properties of this generator, not field rates.
[AI-76] Destination Support Restoration for Finite-Set Multimodal Trajectory Prediction ICRA2027
链接: https://arxiv.org/abs/2609.25942
作者: Fengrui Liu,Jiajun Peng,Duo Peng,Feng Liu
类目: Robotics (cs.RO); Artificial Intelligence (cs.AI)
备注: Submitted to the 2027 IEEE International Conference on Robotics and Automation (ICRA 2027)
Abstract:Robots operating around pedestrians often reason over a finite set of predicted human futures. Repeated online updates can concentrate this limited prediction budget on dominant destinations and leave plausible alternatives underrepresented or absent, removing those alternatives from the finite representation available to downstream decision making. We introduce Destination Support Restoration (DSR), a causal post-selection operator that repairs destination support without retraining the host predictor or increasing the maintained set size. At a repair step, DSR evaluates a temporary destination-stratified candidate bank from the observed prefix, converts candidate evidence into integer target counts, protects representatives of active modes, and reallocates redundant surplus hypotheses to deficient modes. The maintained and returned sets retain exactly N hypotheses, and DSR replaces at most \lceil\rho N\rceil entries. Protected representatives preserve current categorical support; lineage-aware particle filters also preserve surviving resampling ancestors. Each replacement reduces the allocation mismatch to the evidence-driven target by one. On the complete 3,719-trajectory Edinburgh protocol over three seeds, DSR reduces MIF weighted ADE and FDE by 13.36% and 13.30% at N=64 . Paired integrations with CLiFF, PPT, causal GDTS, Social Informer, and PECNet improve both metrics in every evaluated pair. These results show that finite-set support allocation is a useful prediction-side control point when a fixed hypothesis set serves as the interface to downstream systems.
[AI-77] oward Responsible AI-Augmented Cyber Defense: Pattern Recognition Defense-in-Depth and the Case for Human-AI Collaboration
链接: https://arxiv.org/abs/2609.25921
作者: Mustafa S. Aljumaily,Hayder Kareem Abed,Nawar S. Alseelawi
类目: Cryptography and Security (cs.CR); Artificial Intelligence (cs.AI)
备注: 11 pages, 3 figures, 2 tables. Original theoretical and modeling contribution. Simulation code: this https URL
Abstract:Cybersecurity literature has extensively documented the operational benefits of artificial intelligence (AI) for threat detection, incident response, and prevention, while raising qualitative concerns about over-automation, algorithmic bias, and analyst-skill erosion. What remains largely absent is a formal, falsifiable model connecting three constructs that recur across this literature: Defense-in-Depth Theory, the Artificial Intelligence Theory of Pattern Recognition, and human-AI collaboration in security operations. This paper develops such a model. We formalize layered defense as a Bernoulli detection cascade in which AI augmentation enters multiplicatively across layers; we formalize each layer’s pattern-recognition behavior as a Neyman-Pearson/Bayesian detector with a derived closed-form optimal threshold; and we formalize human-AI triage as a capacity-constrained cascade with an explicit, quantifiable trade-off between detection probability and false-alarm (“alert fatigue”) rate. A Monte Carlo/analytical simulation evaluated at illustrative but realistic operating points shows that (i) AI augmentation compounds across defense layers, delivering its largest marginal gains exactly where traditional layering saturates, and (ii) full human review of AI-flagged alerts is not optimal: increasing analyst capacity toward 100% coverage cuts false alarms by roughly 20-fold but simultaneously lowers system-level detection probability, because imperfect analyst accuracy is then applied to every alert rather than a filtered subset. These results give the widely repeated qualitative recommendation of “balanced human-AI collaboration” a precise, testable form and suggest an interior-optimum capacity ratio as a concrete design target for security operations centers (SOCs), including those securing IT/OT-converged critical infrastructure.
[AI-78] Evaluating the Effectiveness of SechKAN on 1D Data
链接: https://arxiv.org/abs/2609.25876
作者: Hoang-Thang Ta
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
备注: 13 pages
Abstract:The connection between the Kolmogorov-Arnold representation theorem (KART) and neural network design has led to the development of Kolmogorov-Arnold Networks (KANs), with applications ranging from STEM problems to AI tasks. In this paper, we investigate the effectiveness of a KAN variant, SechKAN, which relies on hyperbolic secant (sech) functions as basis functions, with a 1D projection to reduce the number of parameters to a level comparable to MLPs. We evaluate SechKAN on three 1D classification datasets: UCI Human Activity Recognition (UCI HAR), ElectricDevices, and Crop, and compare it with several effective networks, including EfficientKAN, MLP, CNN1D, ResNet1D, and DSCNN1D, using approximately comparable parameter budgets. The results indicate that SechKAN achieves competitive performance across the three datasets, with particularly strong performance on Crop. Ablation studies further show that grid size and normalization affect performance, suggesting that SechKAN’s effectiveness depends on the dataset and architectural choices. Our source code and experimental implementation are publicly available at: this https URL.
[AI-79] Agent icSizing: A Large Language Model-based Multi-Agent Framework for Analog Circuit Sizing
链接: https://arxiv.org/abs/2609.25873
作者: Yijia Hao,Pratibha Verma,Dongxu Guo,Cristian Sestito,Michael O’Boyle,Christos-Savvas Bouganis,Themis Prodromakis
类目: Artificial Intelligence (cs.AI); Hardware Architecture (cs.AR)
备注:
Abstract:Analog circuit sizing remains a challenging and time-consuming task due to the large design space, strong performance trade-offs, and increasing circuit complexity in scaled technologies. Although recent large language model (LLM)-based methods show promise in improving sample efficiency and interpretability, existing approaches often lack explicit circuit-topology understanding and are mainly evaluated on relatively simple analog building blocks. This paper presents a multi-agent LLM-based framework for complex analog circuit sizing. The proposed framework first analyzes the circuit topology and decomposes the netlist into functional blocks and substructures. It also extracts lightweight design knowledge for reuse. Based on the extracted topology and knowledge, a planner coordinates multiple role-specialized sizing agents to update design variables and achieve global performance specifications. This workflow mimics the collaborative process of an expert analog design team and provides a structured, interpretable, and simulation-driven optimization procedure. The framework was validated on eight circuits, with the largest design containing up to 55 transistors and 60 sizing variables. Notably, for the LDO benchmark, the proposed method achieved a 60% success rate with an average of 83 iterations, where classical optimizers failed to find feasible solutions. Further, ablation studies demonstrate that topology understanding, design-knowledge infusion, and agent specialization provide complementary benefits. The source code is available to support reproducibility.
[AI-80] Prediction Is Not Detection: Evaluating Pre-Recognition Claims in Longitudinal Clinical AI
链接: https://arxiv.org/abs/2609.25852
作者: Jing Yang,Long R. Jiao,Xiujun Cai,Zongjiu Zhang
类目: Artificial Intelligence (cs.AI)
备注: 27 pages, 1 figure, 3 tables, 1 box; includes Supplementary Note
Abstract:Clinically useful early detection requires validated pre-recognition lead time. Yet event-based evaluations of longitudinal clinical AI can treat recognition-mediated care-process signals as shortcuts and recognition-dependent endpoints as reference standards, inflating apparent performance and lead time while undermining cross-center transport. Such results may serve prognosis without establishing detection before recognition. We define an interval-censored pre-recognition transition, an independent as-of reference standard, and a prespecified recognition proxy to make the claim testable.
[AI-81] Optimizing the Score Losing Sight of the Task: Reward Hacking Across Weights Selection and Prompts
链接: https://arxiv.org/abs/2609.25848
作者: Vansh Wahi
类目: Artificial Intelligence (cs.AI)
备注: 19 pages, 1 figure, 2 tables
Abstract:A higher evaluation score does not always mean a better language model system. When optimization exploits an evaluator’s mistakes, measured progress can conceal unchanged or deteriorating task performance. This failure can arise through parameter updates, selection among generated outputs, or revisions to persistent prompts. We develop a comparative framework for reward hacking across these three optimization substrates: weights, selection, and text. Building on the Proxy Compression Hypothesis and research on inference-time and in-context reward hacking, we examine how reachable behavior, optimization budgets, and persistent adaptation shape exposure to proxy error. We formalize a distance-dependent upper bound on evaluator disagreement and a capacity ordering for nested policy classes, then show why distance alone cannot establish a universal ranking of vulnerability. An exact finite-output illustration demonstrates how the location of a scoring defect changes the behavior favored by each method. We also map representative defenses across substrates, identifying which mechanisms transfer directly and which offer only functional analogies. Persistent prompts receive particular attention: their contents are inspectable, but the behavior induced by a small textual change may be difficult to anticipate. The formal analysis, numerical illustration, and published evidence together provide a basis for comparing optimization methods and identifying the conditions under which their defenses transfer. The resulting framework connects optimization choices to verification requirements: reliable improvement depends on controlling accessible failure modes and preserving evidence of task quality independent of the score being optimized.
[AI-82] In-Context Guidance: Learning Inter-Task Synergies via Numerical Foundational Models for Few-Shot Multitask Optimization
链接: https://arxiv.org/abs/2609.25836
作者: Tingyang Wei,Haofeng Wu,Jiao Liu,Zhao Wei,Puay Siew Tan,Yew-Soon Ong
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Neural and Evolutionary Computing (cs.NE)
备注: In Submission to IEEE Transactions on Evolutionary Computation
Abstract:Multi-task optimization (MTO) addresses a set of optimization tasks simultaneously, often suffering from inaccurate inter-task relationship estimation under limited evaluation budgets, leading to negative transfer. This paper introduces In-Context Guidance Multitask Optimization (ICG-MTO), a novel framework that leverages numerical foundational models to improve inter-task coupling estimation in few-shot scenarios. Unlike conventional methods that rely solely on scarce observed data, ICG-MTO employs a frozen foundational model to infer auxiliary guidance through in-context learning. The framework operates through three stages: constructing an algorithm-specific in-context query from evaluated solutions, using the foundational model to infer a guidance signal characterizing predictive relationships among tasks, and translating this signal into algorithm-specific guidance for maximum-a-posteriori coupling estimation. This approach provides regularization during the early, data-scarce stages of optimization and gradually relinquishes control as task-specific observations accumulate. We instantiate the framework in multitask Bayesian optimization as ICG-MTBO, using directional fitness-class queries to guide inter-task coupling estimation, and further instantiate it in MFEA-II using decision-space-overlap queries to guide random mating probability estimation. Experiments across synthetic benchmarks and a real-world robot arm control problem, together with evaluations under different acquisition functions and evolutionary multitasking, demonstrate the effectiveness and generality of ICG-MTO for few-shot multitask optimization.
[AI-83] CogenPVG: Cognitive-Enhanced Reflective Multi-Agent Framework for Persuasive Video Generation
链接: https://arxiv.org/abs/2609.25821
作者: Yuntian Xiao,Shoulong Zhang,Wenfeng Song,Yan Wang,Yi Chen,Shuai Li
类目: Multimedia (cs.MM); Artificial Intelligence (cs.AI)
备注: 17 pages, 6 figures
Abstract:Persuasive video generation (PVG) is a valuable yet under-explored research topic. Despite the significant advances in multimodal content generation, AI-empowered automated creation of human-made-like videos with substantial persuasiveness remains a formidable challenge. In this paper, we propose CogenPVG, a novel Cognitive-Enhanced reflective multi-agent framework tailored for Persuasive Video Generation task. Given the topic and stance from the user, we decouple the sophisticated generation process into four sequential stages: argument reasoning, storyboard planning, asset creation, and post-editing, imitating the workflow of human video producers. To ensure high persuasiveness, each stage is equipped with a pair of generator and critic agents, following a reflective refinement scheme grounded in a solid psychological theory of persuasion, the Elaboration Likelihood Model (ELM). In the argument reasoning stage, we generate highly logical and credible reasoning thoughts under the guidance of critical thinking theory, enabling cognitive enhancement via the central route of the ELM. For the other three stages, we generate and optimize multimodal assets, assembling them into a persuasive video guided by theories of heuristics, as the peripheral route of the ELM. To the best of our knowledge, CogenPVG is the first work focused on general persuasive topics, without being confined to commercial purposes. Extensive experiments and comprehensive analysis demonstrate that our framework achieves the best persuasion performance, thereby proving the effectiveness of our proposed multi-agent framework for the PVG task.
[AI-84] You Only Need 2/3 of the Chosen Experts: An Empirical Study of Dynamic Expert Pruning in Fine-Grained MoE LLM s
链接: https://arxiv.org/abs/2609.25809
作者: Yuanteng Chen,Qiwei Lai,Chen Tianqi,Peisong Wang,Yuantian Shao,Nanxin Zeng,Zhilei Liu,Chuangyi Li,Jing Liu,Jian Cheng
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
备注: 25 pages, 4 figures
Abstract:Fine-grained mixture-of-experts (MoE) architectures have become a mainstream design for open-weight LLMs, with hundreds of experts and increasingly many selected per token. This shift makes dynamic expert pruning an attractive route to cheaper inference. Yet existing evidence comes largely from coarser architectures and likelihood-scored multiple-choice benchmarks, leaving three central questions open in the fine-grained regime: how redundant per-token expert selection is, how effectively existing pruning methods exploit that redundancy, and what governs a model’s sensitivity to pruning. We fill this gap with a systematic empirical study of twelve fine-grained MoE checkpoints spanning nine architecture families, with a core suite of eleven benchmarks covering knowledge QA, mathematics, code generation, and general reasoning. We find that expert selection is far more redundant than the field’s operating points assume: uniformly retaining about two thirds of the selected experts preserves 98.8% of unpruned performance on average, requiring only a one-integer change and delivering 1.2-1.7x measured speedup across two serving backends. This simple baseline leaves little room for dynamic allocation at conservative budgets: even the best published rules differ from it by under 1% at matched expert budgets. Their value emerges under aggressive pruning, where the best rules recover up to 3.0% over uniform truncation, with gains concentrated in the generative tasks that suffer the sharpest degradation. Sensitivity to aggressive pruning also depends on the model: larger and thinking models are more resilient, whereas multimodal models are more vulnerable. Together, these findings reveal how much expert computation fine-grained MoEs can dispense with, and establish when dynamic allocation earns its complexity, informing both practical deployment and future pruning methods.
[AI-85] When Are Aggregate Agent Traces Diagnosable? Traffic-Governed Interpretation and Calibrated Abstention
链接: https://arxiv.org/abs/2609.25806
作者: Peiying Zhu,Sidi Chang
类目: Artificial Intelligence (cs.AI)
备注: 9 pages, 1 figure, 4 tables. The reproducibility artifact is linked in the paper
Abstract:Runtime traces can appear transparent, but a closed-loop policy determines which states are visited and which failures become visible. We study a simulated hotel-pricing agent mapping time, inventory, and market state to discrete price actions under varying demand regimes. A fault may leave no aggregate trace when the policy rarely visits affected cells. We treat entry into aggregate-only fault interpretation as a diagnosability decision preceding scoring or localization. A reference-map gate requires repeated clean-policy support; a matched runtime gate then requires joint support in clean and current streams. Signal analysis occurs only after both pass. We calibrate false admission on a disjoint clean stream at the physical-component level and model detection by affected clean traffic rather than nominal cell coverage. In a frozen one-shot heldout, 55/72 (76.4%) regime-component units were reference-admitted, representing 20 physical components; 54/55 passed matched runtime admission, while the rejected unit abstained. Stable false admission was 0/20, with a one-sided exact 95% upper bound of 0.1391, meeting the frozen 0.20 criterion. Across 540 repeated unit-arm rows nested in those 20 clusters, affected clean traffic reduced negative log likelihood by 29.3% relative to cell coverage, a gain of 0.1264 nats per row (cluster-bootstrap 95% interval [0.0593, 0.1918]). Adding mask family and its interaction improved log loss by 0.0015 nats per row (one-sided upper bound 0.0066), below the frozen 0.01 practical-sufficiency margin. A development audit found that exact minimum hitting set and greedy selection chose identical supports in 12/12 scenarios because singleton evidence had resolved the conflicts. The result is a bounded rule for interpreting aggregate agent behavior: first establish exposure, then score change, and abstain when the trace cannot support the claim.
[AI-86] he Tasteful Agent : Measuring and Improving Taste in Long-Horizon Tasks
链接: https://arxiv.org/abs/2609.25804
作者: Wenbo Pan,Zhichao Liu,Shujie Liu,Jingying Zeng,Chin-Yew Lin,Xianfeng Tang,Yan Lu,Qi He,Xiaohua Jia
类目: Artificial Intelligence (cs.AI)
备注: 33 pages, 6 figures. Code: this https URL . Dataset: this https URL
Abstract:LLM agents increasingly work on long-horizon tasks, and the decisions they make along the way, such as which hypothesis to test or which implementation to build on, determine the outcome of the whole run. Making these decisions well is becoming a key capability for both engineering and research agents. We refer to the ability to make good long-horizon decisions as the taste of an agent. While existing benchmarks measure the end-to-end success of agents on long-horizon tasks, none of them measures the taste of an agent. To address this problem, we build Taste-Bench, a benchmark of taste questions constructed automatically from trajectories that agents produced in engineering and research tasks. Each question presents a decision fork, a point in a trajectory where multiple directions are available and one of them leads to a better outcome, and the evaluated model chooses among these directions without seeing what happens after the fork. We mine these forks automatically from parallel attempts at the same task and from detours inside a single trajectory, without needing human annotation. We evaluate frontier models on Taste-Bench and find that the best model answers only 59.7% of the questions correctly. We further find that forks whose deciding evidence appears later in the trajectory are much harder for every model, and that a larger reasoning budget does not improve the accuracy. Finally, we show that taste can be trained. We distill the judgment of a teacher that has seen the outcome into a student model, and the student makes better decisions on unseen tasks and improves end-to-end success on held-out SWE-bench Pro tasks.
[AI-87] Evaluating Accuracy and Probabilistic Reliability of Zero-Shot Time Series Foundation Models
链接: https://arxiv.org/abs/2609.25788
作者: Panagiotis Michael,Moysis Symeonides,Demetris Trihinas
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
备注: Accepted for publication at the 30th European Conference on Advances in Databases and Information Systems (ADBIS 2026)
Abstract:Time Series Foundation Models (TSFMs) promise a paradigm shift toward zero-shot forecasting by eliminating task-specific training. However, existing works often overlook trade-offs between predictive accuracy and probabilistic calibration. This paper presents a benchmark study of six TSFMs evaluated on energy, traffic, and financial datasets. We contrast their performance against statistical baselines and a supervised DL model. The study reveals that while TSFMs outperform statistical methods and supervised models, they are subject to a fundamental trade-off between point accuracy and probabilistic reliability. Specifically, xLSTM architectures provide robust probabilistic calibration across horizons. In contrast, patch-based transformers offer competitive accuracy but face calibration issues at long horizons, while transformer-based models exhibit context saturation points for optimal zero-shot reasoning. These findings offer evidence-based guidance for balancing generalization and uncertainty quantification in real-world deployments.
[AI-88] owards Omni-dimensional GUI Agent Navigation with Masked Trajectory Prediction EMNLP2026
链接: https://arxiv.org/abs/2609.25769
作者: Yan Zhang,Pei Fu,Daiqing Wu,Huawen Shen,Ruoceng Zhang,Shaojie Zhang,Jiahui Yang,Yu Zhou,Can Ma,Zhenbo Luo,Jian Luan
类目: Artificial Intelligence (cs.AI)
备注: Accepted to EMNLP 2026
Abstract:Graphical User Interface (GUI) Agents autonomously interact with software to fulfill user requests, where GUI navigation stands out as the most critical and challenging capability. Mastering this capability demands a complex synergy of step-wise decision-making, state-action alignment, and long-horizon planning. While directly mixing these corresponding navigation tasks seems intuitive to simultaneously acquire these skills, such a direct combination is severely bottlenecked by inconsistent optimization objectives and profound data heterogeneity. To overcome these barriers, we propose the MaP (stands for ``\textbfMasked Tr\textbfajectory \textbfPrediction’'), a unified framework that seamlessly harmonizes divergent GUI navigation tasks. By modeling multi-turn GUI interactions as a trajectory and defining training objectives through component masking and prediction, MaP shifts the optimization from task-specific marginal distributions to a consistent objective. Furthermore, to handle the data heterogeneity across multiple navigation tasks, we design a role-aware adapter learning module that dynamically routes each token to a specialized representation space. Extensive experiments on five representative GUI navigation benchmarks demonstrate that MaP effectively mitigates gradient conflicts and significantly outperforms the direct mixture training, establishing a robust paradigm for multi-task GUI navigation.
[AI-89] Neurosymbolic Action Model Learning under Partial Observability
链接: https://arxiv.org/abs/2609.25766
作者: Adem Kikaj,Lennert De Smet,Giuseppe Marra,Luc De Raedt
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:AI planning studies how an agent can reach a goal by executing a sequence of actions. To plan correctly, the agent needs an action model describing when each action can be executed and how it changes the world. Constructing such models by hand requires domain expertise, and can be costly and error-prone. Action models can instead be learned from available data using existing neurosymbolic approaches, but they currently assume access to complete traces of fully observable images . These approaches fail to learn action models under partial observability where some of the images might not be present or are not fully informative of the current state of the world. Hence, this paper proposes NeSyAM, a novel neurosymbolic modeling paradigm for action model learning under partial observability. In addition, the paper presents a unified variational framework for theoretically analysing the limitations of existing methods compared to our proposed approach. NeSyAM is then tested extensively on six visual planning domains and three observation regimes to show it consistently recovers relevant parts of the true action model under partial observability.
[AI-90] he Limits of Simulated Societies: How Post-Training and Survey Fine-Tuning Erase Cross-Cultural Variance
链接: https://arxiv.org/abs/2609.25760
作者: Rojin Ziaei
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Using large language models (LLMs) to simulate diverse human populations has the potential to transform many aspects of computational social science, yet many evaluations score the average response rather than the spread of opinion within real groups. Here, we develop a diagnostic framework that measures point accuracy alongside dispersion retention, the ratio of predicted to human standard deviation ( \dr ), on 10,000 respondent–question pairs from the World Values Survey (WVS) spanning twelve countries and six continents. We evaluate eleven zero-shot language models and five variants fine-tuned on WVS data with SFT, DPO, and GRPO. We identify a failure mode we term \textitconsensus collapse, where alignment training compresses outputs toward one stereotype per group. Along the post-training trajectory from the Llama~3.1 70B base to the Tulu~3 checkpoints, the first stage, supervised instruction tuning, removes half of the spread with minimal accuracy gain ( \dr 1.22 to 0.59; accuracy +0.9 points), the later stages do not restore it, and a gap opens between WEIRD and non-WEIRD countries that survey fine-tuning then deepens while pursuing higher point accuracy. The most accurate model (Tulu~3 70B-DPO fine-tuned on WVS, 57.9%) keeps half the human spread overall ( \dr = 0.50 ) and 11% of it for Nigeria, against 0.70–0.87 for WEIRD countries. Raising the sampling temperature to 1.0 leaves the Wasserstein-1 distance ( \wone ) to human distributions unchanged for both fine-tuned DPO models, and GRPO on Qwen~3.5 9B does not restore the spread under either an accuracy reward or a distribution-shaped reward. Mixing the aligned model with an unaligned prior raises \dr from 0.51 to 0.62 on a held-out split but leaves Nigeria at 0.36. Point accuracy alone therefore misjudges these simulators, and current post-training trades diversity for consensus.
[AI-91] OmniFysics-Nano-V2 Technical Report: Understanding the Physical World Across Modalities
链接: https://arxiv.org/abs/2609.25738
作者: Yizhou Liu,Jinghang Han,Kaixiang Qiu,Qi He,Minghao Han,Yue Jiang,Xujia Chen,Wei Zou,Shunli Wang,Lihua Zhang,Dingkang Yang
类目: Artificial Intelligence (cs.AI)
备注: 24 pages
Abstract:Omni-modal models have expanded multimodal interaction across vision, audio, speech, and language. However, their training is predominantly organized around semantic descriptions and general-purpose objectives, leaving physical attributes, interaction states, and causal mechanisms only partially specified. This gap is not simply a matter of modality coverage: adding more modalities does not by itself provide the supervision needed to connect observations with the physical structure of the world. We present OmniFysics-Nano-V2, a compact omni-modal model for physical-world perception and understanding. The model supports image, video, audio, speech, and text inputs within a shared reasoning framework, together with text and speech generation. To address the lack of explicit physical supervision, we construct a dual-branch physics-aware data pipeline that grounds salient objects in structured physical attributes and aligns visual changes with acoustic events, intermediate responses, and interaction outcomes. To address homogeneous training objectives, we curate reinforcement-learning prompts by reward diversity and adopt a two-stage Group Relative Policy Optimization curriculum that progresses from general task correctness to fine-grained physical perceptual reasoning. Experiments across multimodal, audio-visual, and physical reasoning benchmarks show that the proposed data and training strategy improves physical-world understanding while preserving broad omni-modal competence. The proposed model achieves leading result on 17 of 21 benchmarks against SOTA omni-modal models. By equipping AI systems with both omni-modal and physical-world perception capabilities, OmniFysics-Nano-V2 is poised to become a cornerstone of next-generation Physical AI.
[AI-92] Beyond Class Marginals: Bounding Rehearsal Gaps without Freezing Class Co-occurrence
链接: https://arxiv.org/abs/2609.25735
作者: Congren Dai,Nat Roongjirarat,Fei Ye
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Machine Learning (stat.ML)
备注:
Abstract:Class-balanced replay controls class frequency but does not determine the interval between successive replay appearances of a class. We study this interval, the rehearsal gap, separately from the class marginal and class co-occurrence, and introduce randomised-pass replay (RPR), which visits each resident class once per shuffled pass. For a fixed set of C resident classes and replay batch size b less than or equal to C, RPR preserves the balanced time-averaged class marginal and bounds every gap by 2*ceil(C/b)-1; a churn-conditional bound applies while the resident set changes. The scheduler uses no future class information and adds no replay examples or forward passes. In a linear-head ER-ACE diagnostic, joint absence from the incoming and replay batches produces a one-sided classifier-bias gradient. Longer absence episodes are associated with larger negative bias displacement, and removing the incoming-loss mask attenuates the scheduling effect. In the primary ER-ACE experiments, RPR improves final average accuracy by 0.72-1.67 percentage points relative to independent class-balanced retrieval under reservoir storage, with positive effects also observed under balanced storage. Pretrained ViTs show positive effects on the tested LT10 streams with small replay batches, while matched larger-batch controls show no material effect. Fixed-cycle and reused-pass controls change more than one temporal statistic, so the experiments do not isolate rehearsal-gap length from all other forms of temporal dependence. The accuracy effects depend on the learner and operating regime.
[AI-93] Self-Supervised Combinatorial Optimization with Constraints via Frank-Wolfe
链接: https://arxiv.org/abs/2609.25728
作者: Akbar Rafiey,Yifei Xu,Nikolaos Karalias
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
备注:
Abstract:Self-supervised learning for combinatorial optimization has emerged as a promising paradigm for solving discrete optimization problems with neural networks, but a central challenge remains: handling hard combinatorial constraints within continuous, gradient-based training. Continuously extending combinatorial objectives to convex domains is a powerful technique, yet existing approaches often require projection steps that constrain neural network outputs to lie inside the feasible polytope and rely on ad-hoc and problem-specific constructions. We propose a general framework in which the neural network is allowed to predict arbitrary continuous vectors that could potentially lie outside of the feasible polytope. These predictions are then approximated by sparse convex combinations of feasible solutions using a geometric decomposition algorithm based on Frank–Wolfe methods and approximate Caratheodory results. This decomposition induces an a.e.-differentiable, self-supervised loss defined as the expected value of the discrete objective. The same procedure provides an automatic rounding guarantee at inference time. We demonstrate strong empirical performance across multiple combinatorial problems, including the Quadratic Assignment Problem, Maximum Coverage, and the Traveling Salesperson Problem.
[AI-94] LingLan: An Advancing Traditional Chinese Medicine Diagnosis LLM with Multimodal Data
链接: https://arxiv.org/abs/2609.25715
作者: Zheng Chen,Zhicheng Du,Haoxuan Li,Yingshan Liang,Peiwu Qin
类目: Artificial Intelligence (cs.AI)
备注: 6 pages, 5 figures
Abstract:Though artificial intelligence (AI) increasingly transforms modern medicine, its integration into Traditional Chinese Medicine (TCM) has been relatively slow, primarily due to TCM’s reliance on holistic, subjective diagnostic methods—namely Inspection, Auscultation and Olfaction, Inquiry, and Palpation(I-AOI-P)—which are difficult to align with quantitative, standardized medical systems. In this work, we introduce a Unification Framework for Multimodal Data (UFMD), which automatically processes tongue and pulse images into structured, clinically standard descriptions, integrating multi-source diagnostic information into a unified digital record of I-AOI-P process. Building on this structured data, we create LingLan-14B, a TCM-specific large language model fine-tuned via supervised learning to emulate the diagnostic logic and workflow of I-AOI-P process. Experimental results show that our method significantly enhances diagnostic accuracy, achieving a relative improvement of 103.5% over the baseline (62.72% vs. 30.82%) and reaching an F1-score of up to 82%.
[AI-95] CMaster: Confidence-Aware Querying and Workload-Guided Physical Design for Multi-Source Traditional Chinese Medicine Knowledge Graphs
链接: https://arxiv.org/abs/2609.25712
作者: Zheng Chen,Yuzhu Li,Haoxuan Li,Zhongde Zhang,Lianshun Jin,Peiwu Qin
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Multi-source knowledge graphs (KGs) need query mechanisms that expose reliability and exploit domain structure. This paper presents TCMaster, a property-graph query substrate for confidence-aware traversal and workload-guided physical design over Traditional Chinese Medicine KGs. TCMaster integrates pharmacopoeias, prescriptions, molecular databases, and LLM-extracted micro-semantics into a KG with approximately 221K entities and 723K base edges. It annotates edges with provenance-level confidence, rewrites Cypher queries with confidence predicates, ranks multi-hop paths under PRODUCT, MIN, or weighted-average policies, and uses ontology skew through direction selection, herb-attribute bitmaps, and materialized shortcut edges. On Neo4j, direction selection improves attribute lookup by a factor of 1.47, shortcuts accelerate high-fanout target counting by a factor of 4.42, confidence filtering removes 39.3 percent of low-quality heterogeneous paths, and KG retrieval improves TCMbench QA accuracy by 20.0 percentage points.
[AI-96] oolcompass: Guiding Tool Trialing Not Suppressing It
链接: https://arxiv.org/abs/2609.25678
作者: Junlin Fang,Chong Zhang,Do Nguyen-Thanh,Xiaogang Xu,Zhen Fang,Sean Du
类目: Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
备注:
Abstract:Large language model (LLM) agents must generalize from tools seen during training to unseen tools at deployment. A key challenge is tool trialing, i.e., excessive trials waste the interaction budget, whereas selective trials enable exploration of unfamiliar tools. Existing outcome-based post-training leaves wasteful trials unguided, while turn-level supervision may suppress necessary exploration. We introduce ToolCompass, a post-training framework that guides tool trialing by organizing tool-call representations according to shared functions. Specifically, ToolCompass models each function class as a von Mises–Fisher distribution and jointly reduces intra-function variation across domains and increases inter-function separation. This structure transfers experience from seen tools to functionally similar unseen tools, directing exploration away from unrelated alternatives. ToolCompass requires no ground-truth call traces or unseen-tool access and incurs no inference overhead. Experiments on AppWorld and FTRL show consistent gains across GRPO, RFT, and DMPO. improves AppWorld OOD task success by up to 10.71 percentage points over vanilla post-training and performs best among competitive baselines on both benchmarks.
[AI-97] Seeing Is Not Perceiving: When Synthetic Consumers Can and Cannot Pretest Visual Marketing
链接: https://arxiv.org/abs/2609.25677
作者: Yi-Lin Tsai,Yung-Hsiu(Arvin)Lai
类目: Artificial Intelligence (cs.AI); Computers and Society (cs.CY); General Economics (econ.GN)
备注: 58 pages (30 pages of main text, 23 pages of appendix), 18 figures, 25 tables. All six studies were preregistered on AsPredicted
Abstract:Marketers now deploy generative AI agents as synthetic consumers to pretest visual assets such as logos, packaging, and advertising at a fraction of human-panel cost. However, this procedure assumes that a model seeing a visual cue can also perceive its consumer meaning, which is largely untested. We stress-test the assumption using six canonical visual marketing experiments, varying the two levers managers control: model generation (GPT-4o-mini vs. GPT-5.4-mini) and input format (plain text vs. JSON). Every resulting configuration passed the manipulation checks; however, none of the configurations reproduced more than two of the six human effects, and the remainder were nonsignificant. The one exception was a significant reversal of the human pattern. Providing conceptual or empirical evidence through in-context learning steers average responses toward the human effect. Yet steering has a limit: even when it succeeds, a configuration reproduces less than half of the natural spread of human responses and so understates consumer heterogeneity. We integrate these results into an AI governance protocol (Calibrate, Intervene, Deploy) that delineates when synthetic consumers can responsibly screen creatives and when human panels remain necessary.
[AI-98] aching Reinforcement Learning and Humanoid Robotics to High-School Students: An Expert-Validated Curriculum Design on a Low-Cost Open Platform
链接: https://arxiv.org/abs/2609.25674
作者: Yuanzhe Dong,Jie Cao,Shuman Wang
类目: Robotics (cs.RO); Artificial Intelligence (cs.AI)
备注: 6 pages, 4 figures, 4 tables. Accepted for presentation at the IEEE International Conference on Teaching, Assessment, and Learning for Engineering (TALE 2026)
Abstract:Lower cost open source robots and reinforcement learning (RL) simulation tools create new opportunities for precollege students to engage with contemporary robotics. However, translating a complete research workflow, spanning mechanical assembly, electrical setup, simulation, policy learning, system identification, and physical deployment, into a coherent course for novice learners remains challenging. We present an integrated robotics course framework that organizes these activities around a shared robotic artifact. The framework combines parallel disciplinary tracks, sequencing based on technical dependencies, progressive integration of simulation and hardware, layered performance checkpoints, and structures for balancing collaborative work with individual accountability. We illustrate the framework through a high school curriculum organized around a robot project in which pairs of students assemble an open source humanoid robot, train a walking policy in simulation, and deploy it on the physical platform. The framework was developed through an iterative design process that included formative review by five experts in robotics research, engineering, secondary STEM education, and curriculum design. Expert feedback highlighted three central design tensions: authenticity versus cognitive load, system integration versus timely visible progress, and team construction versus individual accountability. These tensions informed the final framework presented in this paper. This work offers a structured approach for adapting robotics research workflows into interdisciplinary precollege courses; future classroom studies are needed to examine implementation and student learning.
[AI-99] From Experts to Sub-experts: Fine-grained Parameter-Efficient Fine-Tuning for MoE LLM s
链接: https://arxiv.org/abs/2609.25655
作者: Zhentao Tan,Chang Liu,Yao Liu,Yue Wu,Jieping Ye
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
备注:
Abstract:As large language models (LLMs) scale rapidly, dense full-parameter adaptation becomes increasingly expensive, motivating sparse and modular architectures such as Mixture-of-Experts (MoE) models. This shift raises a key question for parameter-efficient fine-tuning (PEFT): at what granularity should parameters be selected and updated? Existing PEFT methods such as LoRA operate on predefined weight matrices, while expert-level sparse tuning methods update entire selected experts. However, we observe that activated experts are internally sparse, with only a small fraction of intermediate channels strongly responding to downstream tasks, indicating that expert-level adaptation is still too coarse. We propose NSFT (Neural Sub-expert Fine-Tuning), a fine-grained PEFT framework that refines MoE adaptation from experts to sub-experts. NSFT decomposes each expert along the intermediate dimension into structured channel groups and selects task-relevant sub-experts by combining routing importance with intra-expert activation saliency. To optimize sparse partial updates, NSFT further introduces learning-rate scaling and dynamic gradient scaling to compensate for the reduced effective update magnitude. Experiments on OLMoE and Ling-mini-2.0 across challenging domain-specific tasks and general benchmarks show that NSFT consistently outperforms representative PEFT and expert-level sparse tuning baselines, while using substantially fewer trainable parameters and preserving competitive general capability. These results suggest that sub-expert-level adaptation is a more precise and efficient PEFT paradigm for MoE LLMs.
[AI-100] sting-Driven Reliability Audit of Trajectory-Based Early Outcome Prediction for LLM Agents : Target-Specific Calibration Transfer Persists Within a Single Benchmark
链接: https://arxiv.org/abs/2609.25647
作者: YanZe Cao
类目: Artificial Intelligence (cs.AI)
备注: 26 pages, 4 figures, 3 tables
Abstract:Predicting early outcomes based on trajectory can decrease the expenses associated with agent evaluation by terminating a run once the outcome becomes sufficiently predictable, assuming that the predictor’s confidence is properly calibrated. Calibration is at risk when a predictor is applied to an agent on which it was never trained, but it is not known whether such transfer failures are broad across agent systems or concentrated in specific target agent/head combinations. Using public SWE-bench Verified trajectories and a frozen dual-head early-outcome prediction pipeline, we ran a leave-one-agent-out calibration audit, a shared-predictor leave-two-agents-out control, oracle prior correction, and a robustness battery over training cohorts, task resampling, task halves, jackknife, and thresholds. Fixed-scaffold TerminalBench analysis served as a pre-registered boundary test. Broad same-predictor pairwise heterogeneity was not supported; the median pairwise corrected-gap differences were 0.0180 (SUCCESS head, 45 pairs) and 0.0385 (FAILURE head, 35 pairs), and the pre-registered heterogeneity criterion was not met on either head. Two specific combinations, gpt-5-mini/SUCCESS and claude-opus-4.6/FAILURE, showed persistent calibration-transfer errors (median corrected gaps 0.1377 and 0.1107) without a sign reversal under any frozen control. TerminalBench did not establish cross-benchmark replication: the success target produced zero decisions (INDETERMINATE), and the failure target did not satisfy the pre-registered persistence criterion. Therefore, a strong target-specific calibration-transfer error can exist within one frozen environment, but the evidence does not establish that the error is intrinsic to the model or general across benchmarks.
[AI-101] Ladders of Thought: A Self-Evolving Curriculum of Progressively Simplified Reasoning Traces
链接: https://arxiv.org/abs/2609.25643
作者: Minghui Liu,Thomas Magelinski,Dehao Yuan,Qi Yu,Furong Huang
类目: Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
备注:
Abstract:Large language models (LLMs) excel at reasoning when scaled to hundreds of billions of parameters, but small- and mid-scale models remain brittle reasoners even with knowledge distillation (KD). We present Ladders-of-Thought (LoT), a framework that improves reasoning by combining progressive question rewrites with a self-evolving curriculum. LoT automatically generates semantically faithful but easier variants of reasoning problems, organizes them into difficulty buckets using step-based measures, and employs a self-evolving bandit scheduler to allocate training adaptively. Evaluated on two reasoning domains, math and multi-hop reasoning, across 1-8B models from different families, LoT consistently improves over KD. It delivers large gains on arithmetic tasks (e.g., +32 percentage points on AddSub, +25pp on SVAMP), +2-8pp improvements on in-domain test splits, and strong though dataset-dependent benefits on multi-hop reasoning (e.g., +16pp on QASC, +25pp on StrategyQA). LoT also converges faster than staged curricula, highlighting the value of adaptive progression. These results show that progressive rewrites coupled with adaptive curricula provide a simple yet effective recipe for strengthening reasoning in smaller LLMs.
[AI-102] An Exploratory Replica-Overlap Probe of the Grokking Transition
链接: https://arxiv.org/abs/2609.25634
作者: A. C. Opus,J. Q. Lu
类目: Machine Learning (cs.LG); Disordered Systems and Neural Networks (cond-mat.dis-nn); Artificial Intelligence (cs.AI)
备注: 13 pages, 5 figures. Pre-registered study; the registered decision rule returns UNDETERMINED (reason code C0_INSTRUMENT_INVALID) because the registered alignment implementation is not function-preserving. Reported as an honest negative/instrument-invalid result
Abstract:We trained 64 independently seeded networks in four configurations, continuing each to sustained convergence or a 40,000-epoch ceiling. We then asked whether an RSB-inspired distribution of pairwise weight overlaps changes across the grokking transition. It is the alignment step, not the overlap statistic, that determines what this registered probe can report. The registered implementation permutes hidden units without the corresponding bias and head-internal permutations and therefore does not preserve the network function. Every q_wt value computed through this alignment inherits the defect; q_fn does not, because it is computed from predictions of the unpermuted models. The numerical-precision requirement also failed, and an audit found protocol deviations. Consequently, the pre-registered rule gives no verdict: registered outcome UNDETERMINED (reason code C0_INSTRUMENT_INVALID). These data provide neither a confirmatory null nor a validated reading of the Parisi order parameter. Only frac40 cleared the 12/16 checkpoint-completeness requirement. For this configuration, a post-hoc criterion applied to the same data gave a Hartigan-dip interval containing zero (95% CI for Delta dip = [-0.017, 0.034]), whereas the overlap standard deviation increased by a factor of about 5.6. A post-hoc calibration assigns the dip test zero power at the simulated separations; the interval is therefore uninformative, not evidence of no change. The standard-deviation ratio is the only statistic here with power at the observed effect. Ensemble loss was near-flat only under the pre-specified 1% threshold. Finally, grokking rates of 0/16, 11/16 and 16/16 remain descriptive because train fraction is confounded with split identity.
[AI-103] What Should a Self-Teacher See? Privileged Context Design for On-Policy Self-Distillation
链接: https://arxiv.org/abs/2609.25623
作者: Kanghui Tian,Siyuan Liu,Tianxiang Jiang,Shuai Dong,Yizhuo Li,Tian Ding,Yuan Guo,Songze Li,Haowen Hou,Congcong Wang,Yi Wang
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
备注:
Abstract:More privileged information does not always make a better teacher. We study this tension in on-policy self-distillation (OPSD), where a frozen copy of the base model scores the student’s own rollouts under privileged context, conventionally a complete reference solution that bundles the final answer with one particular reasoning path. Holding the student view and training fixed within each scale, we compare that default against three abstractions compiled offline, a named strategy, a method-independent framing, and a problem category, and against an answer-only control that keeps the destination but removes the path. In the primary runs on competition mathematics, the best intermediate contexts improve the in-domain peak mean over the full solution by 1.4 points at 4B and 1.6 at 8B, while storing an order of magnitude fewer hint tokens. Comparisons across three seeds also show positive mean gains for the framing and category contexts at both scales. Answer-only conditioning remains competitive in the primary runs, within 0.2 points of the full solution at these scales. The preferred context varies with student scale and task. Initial teacher-student KL does not order downstream performance. What a self-teacher should see is therefore not everything it could, but the level of abstraction its student can still act on.
[AI-104] ChatT2: An Adaptive Framework for Developing a Large Language Model-Based Agent for Natural Product Domain Research
链接: https://arxiv.org/abs/2609.25620
作者: Yihan Wang,Qiandi Gao,Yihui Zhuang,Liangjun Ge,Heqian Zhang,Jiaquan Huang,Zhiwei Qin
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Scientific investigations into microbial natural products (NPs) present significant challenges for novices, largely due to the complexity of microbial systems, biochemical diversity, technical skill requirements, and the demands of bioinformatics and data analysis processes. To address these issues, we introduce ChatT2, a large language model (LLM)-based agent that is specifically tailored to the unique characteristics of bacterial type II polyketides. These polyketides form a structurally distinct and therapeutically important NP family. ChatT2 was developed within an autonomous multiagent framework composed of a mentor, an executor, and an evaluator, each with defined responsibilities. The mentor acts as an intermediary between ChatT2 and the user, utilizing chain-of-thought prompting to refine the intent of the user. Under the guidance of the mentor, the executor synthesizes multimodal information via retrieval-augmented generation techniques and seamlessly integrates bioinformatics and cheminformatics tools. The evaluator ultimately assesses the output of the executor to ensure the richness and accuracy of the retrieved information. Our research highlights how ChatT2, designed with this multiagent framework, addresses the challenges faced by general LLMs in terms of understanding limited, specialized corpora and complex biological information and provides both experts and novices with a valuable tool for exploring various NPs of interest. The ChatT2 webserver can be accessed at this https URL.
[AI-105] Reasoning -Preserving Fine-Tuning of Post-RL LLM s with Null-Basis LoRA
链接: https://arxiv.org/abs/2609.25618
作者: Wenzhi Fang,Nicholas Tzou,Lazar Valkov,Srinivas Chappidi
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Reinforcement learning (RL)-based post-training has become an effective approach for eliciting reasoning capabilities in large language models (LLMs). However, adapting post-RL models to new knowledge domains or behaviors through subsequent supervised fine-tuning (SFT) can severely overwrite these capabilities. Existing approaches mitigate such forgetting through experience replay, specialized initialization, or constrained optimization using gradient projection, but either provide limited preservation or incur substantial training overhead. Our analysis shows that reasoning activations concentrate in low-dimensional subspaces, leaving substantial null-space capacity for adaptation, and that the corresponding approximate null spaces can be reliably estimated from a modest number of examples. Motivated by these observations, we propose Null-Basis Low-Rank Adaptation (NB-LoRA), a parameter-efficient method for adapting post-RL LLMs while preserving their acquired reasoning ability. We formulate reasoning retention as a layer-wise hidden-state preservation constraint and construct a fixed approximate null basis from reasoning activations. LoRA updates are then reparameterized through this basis, enforcing the preservation constraint throughout fine-tuning. Extensive experiments across multiple RL-trained LLMs and diverse downstream tasks show that NB-LoRA matches standard LoRA in adaptation performance, maintains reasoning accuracy near pre-fine-tuning levels, and generalizes this preservation to held-out reasoning benchmarks.
[AI-106] ArticleMiner: Ontology-Guided Knowledge Graph Construction from Scientific Publications ISWC2026
链接: https://arxiv.org/abs/2609.25607
作者: Md Abrar Jahin,Craig A. Knoblock,Jay Pujara
类目: Artificial Intelligence (cs.AI)
备注: Sci-K co-located with ISWC 2026, Bari (Italy)
Abstract:Scientific papers keep much of their quantitative content in tables and supplementary files, where a number means something only through its header, caption, unit, analytical method, and the conventions of its field. Recovering the rows and columns of a table is therefore not the same as recovering the scientific fact it reports. Most semantic table-interpretation methods assume that a clean table is already available and subsequently map its cells or columns to ontology terms, whereas most publication-level extraction systems are designed for a single domain. We study a middle path: a shared process that reads a paper and its supplementary files, gathers evidence from several parsers and a language model, and reconciles that evidence, while a bounded human-authored task module for each task supplies the domain meaning. The module lists the canonical names the graph may use, the surface forms that map to them, a small set of derivation rules and validity constraints, an identity key, and the bindings used to write RDF. It defines what a task is allowed to emit; it does not try to list every convention of a field. We build four such modules (for drug-discovery chemistry, materials science, machine learning, and mineral geochemistry) in the ArticleMiner framework, and evaluate them on 163 papers, including a new geochemistry benchmark with expert-curated ground truth. In comparisons against a same-LLM few-shot baseline, the point estimates favor ArticleMiner on all four tasks, with uncertainty on the two smaller benchmarks. The geochemistry comparison also includes access to supplementary files, so its improvement cannot be attributed to domain guidance alone.
[AI-107] Evaluating Coding Agents on Kernel Exploit Generation
链接: https://arxiv.org/abs/2609.25591
作者: Junyoung Jang,Gwanhyun Lee,Hwiwon Lee,Kyuheon Kim,Jongseong Kim,Jinho Jung,Lingming Zhang
类目: Artificial Intelligence (cs.AI); Cryptography and Security (cs.CR)
备注:
Abstract:Coding agents now find real vulnerabilities in production software. However, bug discovery results do not measure whether agents can construct exploit primitives. We introduce KEX-bench, a benchmark for evaluating coding agents on exploit primitive generation against real operating-system kernels. KEX-bench contains 45 task instances across 40 Linux and Windows CVEs, covering kernel address leak, instruction-pointer control, heap read, heap write, and arbitrary address write. Each task runs in an isolated virtual machine, exposes controlled tools, and uses a deterministic verifier to check primitive-specific success. We evaluate state-of-the-art coding agents paired with frontier and open-weight models under fixed tool-call budgets. Without a reference proof of concept (PoC), the strongest configuration solves 1 of 20 Windows tasks (5.0%) and 14 of 25 Linux tasks (56.0%). With a reference PoC, the strongest configuration solves 31 of 45 tasks (68.9%). This highlights the gap where agents reach kernel crashes but fail to shape kernel state into exploit primitives. We release KEX-bench for reproducible research on AI-assisted exploitation at this https URL.
[AI-108] ransformer Heads Looking for Order
链接: https://arxiv.org/abs/2609.25588
作者: Jasper van Doornmalen,Alexander Kozachinskiy,Corinna Mathwieser,Tomasz Steifer,Felipe Urrutia,José Verschae,Przemysław Andrzej Wałȩga
类目: Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
备注:
Abstract:In this note, we show that the problem of checking, whether a sequence of bits is ordered, is not doable by 1-head 1-layer transformers but is doable by a 2-head 1-layer transformer. Unlike similar previous results, our results assume the model where transformers have an output MLP.
[AI-109] EMGBlend: Heterogeneity-Aware Self-Supervised Pretraining for Gesture and Force Decoding
链接: https://arxiv.org/abs/2609.25582
作者: Yuwei Jia,Cheng Zhong,Jinyang Yu,Zhe Cui
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
备注:
Abstract:Public surface electromyography (EMG) datasets vary widely in electrode layout, channel count, frequency support, and size. Simply mixing them for pretraining can misalign channel semantics, introduce spectral targets that some devices cannot observe, and let large or high-channel-count datasets dominate learning. We introduce EMGBlend, a self-supervised framework designed around these differences. It combines shared channel patches with geometry-aware attention, restricts spectral targets to each recording’s supported frequency band, and balances exposure across data sources. We pretrain a 109M-parameter model on 11 public EMG sources and evaluate it on gesture recognition, continuous-force regression, and contact classification. EMGBlend consistently outperforms matched random initialization and waveform reconstruction controls. Fixed-budget source controls show that multi-source pretraining improves gesture recognition and remains competitive for force decoding. Ablations confirm that geometry, band-aware targets, and source balancing each contribute to transfer, although cross-person NinaPro force estimation remains difficult. Overall, EMGBlend shows how heterogeneous EMG datasets can be combined through explicit mechanism design rather than simple concatenation. Code is available at this https URL
[AI-110] Gaze responses to false-positive computer-aided detection prompts during colonoscopy: a paired-video and real-time eye-tracking study
链接: https://arxiv.org/abs/2609.25581
作者: Te Luo,Yan Zhu,Peiyao Fu,Ruijie Yang,Xian Yang,Quanlin Li,Pinghong Zhou,Shuo Wang
类目: Artificial Intelligence (cs.AI)
备注: 21 pages,3 figures
Abstract:False-positive computer-aided detection (CADe) prompts may divert endoscopists’ attention during colonoscopy, yet the attentional impact of individual prompts remains unclear. We used event-locked eye tracking to quantify gaze attraction and attention occupation in complementary retrospective and prospective studies. In a retrospective paired-video experiment, 3 senior and 2 novice endoscopists viewed 60 colonoscopy videos with and without CADe. The prospective study recorded gaze during 42 real-time CADe-assisted colonoscopies performed by 9 senior endoscopists. Screened CADe prompts outside expert-annotated lesion windows were classified as false-positive artifact events. False-positive prompts attracted gaze in 48.6% (68/140) of retrospective observations and 65.2% (533/817) of prospective events. Among attraction events with complete recovery, median attention occupation lasted 1000 ms in the retrospective study and 1100 ms in the prospective study. Corresponding median prompt durations were 33 ms and 267 ms, with median time amplifications of 17.55-fold and 5.15-fold, respectively. In paired retrospective comparisons, visible artifact prompts drew gaze closer to the prompted region than did the same-coordinate unassisted reference. Secondary retrospective analyses showed high lesion gaze recognition without and with CADe (98.0% versus 99.0%). First gaze entry into lesion regions occurred 147.8 ms earlier with CADe. Across controlled and real-time clinical settings, false-positive CADe prompts frequently captured gaze, with attention persisting beyond prompt visibility. These findings support considering prompt-related attentional burden in CADe evaluation and design.
[AI-111] Direct Optimization of Generators for Search in Automated Theorem Proving
链接: https://arxiv.org/abs/2609.25575
作者: Adam Ousherovitch,Ambuj Tewari
类目: Artificial Intelligence (cs.AI); Machine Learning (stat.ML)
备注:
Abstract:Fine-tuned Large Language Models (LLMs) significantly advance Automated Theorem Proving (ATP), but are often deployed as guiding policies within tree search rather than for single-attempt generation. Recent work shows cross entropy is suboptimal for an LLM used in flat search strategies such as aggregation or filtering and that work has developed new loss functions to correct this misalignment. Extending this alignment to tree search is more challenging: proof discovery depends on exploration and recovery through off-trace states that supervised demonstrations do not reveal. We extend Compute-Aligned Training (CAT) to this setting through an abstraction of policy-guided search, deriving tractable, trace-supported losses. Alongside these search-aware losses, we introduce a search-agnostic uniform-allocation (UA) loss that accounts for the budget without specifying the specific search. Both induce scalar weights on per-tactic cross-entropy gradients. We characterize how off-trace behavior affects the search-aware weights, including conditions for vanishing approximation error at large budgets. On a Lean benchmark, both approaches achieve higher observed proof-success rates than cross-entropy across six search strategies, with strong results from a single shared UA adapter. Budget sweeps show larger gains over cross-entropy at 16 than at 256 expansions, implying CAT scales with test time compute.
[AI-112] A Behavioral Trait Leaks into Preferences: Diagnosing Trait Interference in LLM User Simulators CIKM2026
链接: https://arxiv.org/abs/2609.25572
作者: Chaehyun Kim,Sein Kim,Hongseok Kang,Chanyoung Park
类目: Artificial Intelligence (cs.AI)
备注: CIKM 2026 short
Abstract:LLM-based user simulators aim to bridge the offline-online gap in recommender evaluation by emulating users through injected traits, where preference attributes determine what a user engages with and a behavioral activity trait governs how long they browse. However, we show this intended trait independence collapses during simulation, causing two failures: (i) Trait Interference, where amplified activity distorts preference boundaries and forces interactions with mismatched items to sustain browsing, and (ii) Evaluation Invalidity, where satisfaction scores inflate with activity-driven page counts despite taste mismatches, biasing evaluation toward trait distributions rather than recommender performance. To resolve this, we propose PQA, a page-level quality anchoring method that guides simulators using a personalized anchor reflecting each user’s intrinsic preference standard. By assessing whether a page meets this standard before further browsing, PQA enables proactive exits from low-quality pages, letting the activity trait retain its intended role of modulating browsing depth within preference-conforming pages. Experiments show PQA mitigates trait interference and improves the reliability of LLM-based simulator evaluation under activity shifts. Our code is available at this https URL
[AI-113] Recovering Agent ic Sovereignty: Mitigating the Consensus Paradox via Contrastive Epistemic Decoding
链接: https://arxiv.org/abs/2609.25570
作者: Dahlia Shehata,Ming Li
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Large language models (LLMs) exhibit a parametric vulnerability to adversarial swarm consensus. To mitigate this sycophancy, we introduce Contrastive Epistemic Decoding (CED), a zero-shot inference intervention. Unlike standard Contrastive Decoding (CD) which relies on a weaker secondary model, CED utilizes a dual forward-pass on a single architecture to isolate conformity bias. By introducing a novel asymmetric, zero-bounded probability clamp and discrete top-k truncation mask, CED mathematically suppresses toxic consensus tokens without causing grammatical collapse. Evaluated across 7,200 paired trajectories on complex benchmarks (GAIA, SWE-bench, Multi-Challenge) using Gemma-2 (9B), Llama-3.1 (8B), and Mistral v0.3 (7B), CED successfully neutralizes architectural and positional biases. By reducing cognitive loafing by up to 33.00% absolute, CED drives significant performance gains, yielding up to a 30.75% accuracy recovery. Regaining sovereignty induces distinct architectural behaviors—passive task-focus in Gemma-2 and active refutation of the simulated swarm in Llama-3.1—showing CED decouples compliance from capability without fine-tuning.
[AI-114] AkasicMEM: Governed Enterprise Memory for Agents
链接: https://arxiv.org/abs/2609.25563
作者: Jeongmin Bae,Yongjae Kim,Kyoung Hur,Donghyoung Han,Min-Soo Kim
类目: Databases (cs.DB); Artificial Intelligence (cs.AI)
备注: 9 pages, 3 figures
Abstract:Agent memory enables enterprise agents to retain knowledge acquired during work and reuse it across tasks and agents, turning execution experience into persistent organizational knowledge. Realizing this potential requires both source–memory integration, through which enterprise sources and accumulated memory can be utilized together, and memory governance, through which shared memory remains subject to organizational policies throughout its lifecycle. These requirements interact when information from enterprise sources persists in memory. As this information is repeatedly derived and reused under changing principals and policies, source restrictions may be bypassed, resulting in information leakage. Preventing such leakage requires authorization continuity, under which source restrictions remain effective throughout source-to-memory and memory-to-memory derivation and reuse. Existing approaches address these concerns individually, but do not treat source–memory integration, memory governance, and authorization continuity as combined core design targets across the memory lifecycle. We define Governed Enterprise Memory as agent memory designed around this combined scope and present AkasicMEM as its realization. AkasicMEM realizes authorization continuity through transitive lineage, policy composition during memory formation, and policy re-evaluation during retrieval. It is built on GraphAI’s AkasicDB, a unified vector–graph–relational database whose storage and execution substrate enables the underlying operations of these mechanisms to be jointly optimized and executed.
[AI-115] IndustrialVLA-Bench: A Traceable Multi-Axis Evaluation of Open Robot Policy Models
链接: https://arxiv.org/abs/2609.25562
作者: Yiqi Wang,Zhifeng Rao,Jiaqi Zhang,Xiaoyang Li,Zhangkai Wu,Yiqun Duan,Mingkai Zheng,Fei Wang,Shan You,Taotao Cai
类目: Robotics (cs.RO); Artificial Intelligence (cs.AI)
备注: preprint
Abstract:Open robot policies increasingly follow two paradigms: vision-language-action models (VLAs) directly map observations and instructions to actions, whereas world-action models (WAMs) incorporate learned video or world dynamics into policy learning or action generation. Although both target the same manipulation tasks and represent alternative design choices, they are commonly reported under different evaluation protocols, leaving their capability, robustness, language sensitivity, and deployment-cost trade-offs unclear. We present IndustrialVLA-Bench, an evidence-aware evaluation of six released VLA and WAM systems under a unified reporting schema. It separately evaluates clean capability on LIBERO, non-language robustness on LIBERO-Plus, instruction sensitivity on LIBERO-Para, and observed execution cost. Reported task scores aggregate three complete evaluations with distinct random seeds under a fixed checkpoint and inference configuration. Across all six systems, clean LIBERO averages differ by only 1.58 points, whereas robustness and paraphrase summaries span 14.62 and 31.08 points. Restricting every comparison to the three protocol-faithful systems preserves the effect (1.36, 14.62 and 23.10 points), so the diagnostic separation reported here does not depend on the weaker evidence tiers. We additionally report observed inference latency, peak memory, runtime mode, and an evidence status for every system. Protocol-faithful, near-reproduction, and pending-verification entries remain visibly separated; only protocol-faithful entries support strict comparisons. Rather than claiming universal superiority of either paradigm, IndustrialVLA-Bench provides traceable evidence for comparing released robot policies on shared practical criteria. Code and evaluation records are available at this https URL.
[AI-116] Weakly Supervised Quantum Error Mitigation
链接: https://arxiv.org/abs/2609.25555
作者: Seyed Mohamad Ali Tousi,G. N. DeSouza
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Supervised approaches to quantum error mitigation learn a map from noisy circuit outputs to ideal ones, and therefore require the ideal outputs. Producing those ideal outputs demands noiseless classical simulation, whose cost grows exponentially with system size, so supervision is unavailable in exactly the regime where mitigation matters most. We ask whether cheap, individually unreliable signals drawn from circuit structure and hardware calibration can take the place of ideal labels. We assemble sixteen heuristic labeling functions (stabilizer and parity constraints, relaxation and readout characteristics, local depth, gate counts, and neighboring activity), reconcile their disagreements with a probabilistic label model, and read the resulting per-qubit error probabilities as a readout channel whose inverse mitigates the measured distribution. No ideal output enters the training path. On 147,000 five-qubit circuits executed on two IBM devices, the method removes 24.3% (Algiers) and 28.8% (Hanoi) of the Kullback-Leibler divergence to the ideal distribution, against 15.4% and 21.5% for the strongest published analytical baseline, a margin that holds on both devices and lies far outside its bootstrap interval. Supervised neural models trained on ideal distributions remain stronger where such labels exist, and we quantify that gap rather than setting it aside; the method’s claim is to the regime where they do not, since the labels they require cannot be computed for the circuits mitigation is needed for. The codes will be released shortly.
[AI-117] DefaultGNN: A Dual-Perspective GNN Framework for Predicting Corporate Default from Buyer-Seller Transaction Networks CIKM2026
链接: https://arxiv.org/abs/2609.25542
作者: Junghoon Kim,Hyunsung Kim,Seungyoon Choi,KyoungYong Park,Jihun Lee,YongGu Ji,Chanyoung Park
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
备注: CIKM 2026 Applied Research Track
Abstract:Corporate default prediction is a core problem in financial risk management, yet traditional credit models rely heavily on financial statements that are often sparse or unavailable for many firms. Corporate transaction networks offer a complementary view of real economic activity, but how risk propagates through buyer-seller relationships remains underexplored. We conduct a large-scale empirical study using real-world electronic tax-invoice data spanning six years that links transaction histories with default events, revealing that transaction-driven risk is both role-dependent (buyer or seller) and scale-dependent. Based on these findings, we construct multiplex buyer-view and seller-view transaction networks and propose DefaultGNN, a dual-perspective graph neural network-based framework for corporate default prediction. DefaultGNN integrates both views to model how risk flows through transactional relationships, achieving strong improvements over both attribute-based and graph-based baselines, especially for firms with limited intrinsic risk signals. We further provide interpretable network-based explanations by visualizing how distressed trading partners contribute to default risk. In collaboration with a licensed credit rating agency, we validate that DefaultGNN’s predictions complement existing credit scoring models, improving approval rates by 7-11%p without increasing default risk among approved firms. The source code can be found at this https URL
[AI-118] A JEPA Recipe for Tabular Foundation Models
链接: https://arxiv.org/abs/2609.25541
作者: Mingyu Jeon,Suwan Cho,Jae Young Suh
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
备注: 16 pages, 5 figures
Abstract:Tabular foundation models learn to predict cell values in context, whereas world-model self-supervision asks for prediction in representation space (LeCun, 2022; Assran et al., 2023). On a tabular foundation-model prior, the latent term of a joint-embedding predictive architecture (JEPA) collapsed in our earlier runs and took the encoder with it to a constant map. We report a recipe under which the latent term survives to convergence beside the value objective: the value head reads the encoder field rather than the predictor, and the target is an exponential moving average (EMA) difference. To bound its cost against the value-only arm, both arms train until a plateau rule stops them, with no fixed step budget. A fixed horizon had confounded a slowdown with a ceiling, since the value-only arm was still improving well past the usual budget. At convergence, in one run per arm, the JEPA arm trails the value-only arm across 147 real datasets, 32:70 wins to losses on classification (29:63 with one entry per dataset name) and 8:24 on regression, the margin small on classification and wider on regression, and the count leans the same way in each stratum and each benchmark. The JEPA arm (jepa) needs 1.42 times as many steps as the value-only arm (ds), and 1.66 times its wall-clock, to reach its plateau.
[AI-119] Hill Sampling for Test-Time Scaling: A Simple and Better Alternative to Repeated Sampling Evolution and Training
链接: https://arxiv.org/abs/2609.25510
作者: Jacob Beck,Philip V. Ogren,Ari Kobren
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
备注:
Abstract:Large language models (LLMs) can improve solutions to verifiable scientific and algorithmic problems by spending additional computation at test time. Recent systems achieve strong results with increasingly elaborate evolutionary search harnesses or by updating model parameters during test-time training. We ask how much of this machinery is necessary. We introduce Hill Sampling, a simple procedure that repeatedly samples candidate program edits from a frozen LLM, retains the best program found so far, and conditions all subsequent samples on that program. We evaluate the method on circle packing, sums/differences of sets, and Erdos’ minimum-overlap problem using three open-weight models. Hill Sampling sets a new state of the art on circle packing among published methods, improves over the AlphaEvolve reference on Erdos’ minimum-overlap problem, and achieves strong results on sums and differences of finite sets. The circle-packing and Erdos results require only hours of wall-clock time on eight NVIDIA H100 GPUs. To our knowledge, we also conduct, the largest study, by parameter count, of evolution strategies (ES) applied directly to LLM weights at test time. Surprisingly, learning the weights is worse than setting the ES learning rate to zero: at zero learning rate, the method is still searching in weight space through fixed random perturbations. Those perturbations can help exploration, but randomness from token sampling is stronger still, and repeated sampling remains substantially weaker than Hill Sampling. These results suggest a simple test-time compute allocation strategy: repeatedly sample edits to the best verified solution found so far, before introducing additional complexity such as adding archives, diversity mechanisms, evolutionary scaffolds, or test-time parameter learning.
[AI-120] SMTB: Fast Structure-Mapping with Tight Bounds
链接: https://arxiv.org/abs/2609.25508
作者: Daniel Weitekamp,Christopher MacLellan
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Structure-mapping forms analogies by aligning systems of relationally connected elements based on shared structure instead of surface features. We introduce a new structure-mapping algorithm: Structure-Mapping with Tight Bounds (SMTB) that is 5–15x faster than the structure-mapping engine (SME) and about 50% better at finding mappings in large nested domains. SMTB is part of the broader Cognitive Rule Engine (CRE) project, a flexible multi-language-compatible framework with an accessible Python interface to state-of-the-art C++ implementations of core algorithms commonly used in cognitive systems such as pattern matching, planning, and structure-mapping. CRE and SMTB are designed to work with a wide range of representation choices. Unlike SME, which biases higher-order correspondences in tree-like predicate logic, SMTB maximizes relational connectivity without privileging higher-order relations. This allows SMTB to work just as well over arbitrary relational graphs as it does in tree-like domains of nested predicate logic. We discuss situations where privileging “higher-orderness” in structure-mapping can cause issues, and illustrate how SMTB avoids failure modes that SME would encounter in these situations. We also provide an evaluation comparing SMTB to SME v4 over 5845 domain pairs from the SME corpus.
[AI-121] owards participatory speech dataset curation: A queer case study and conceptual framework INTERSPEECH2026
链接: https://arxiv.org/abs/2609.25496
作者: Brooklyn Sheppard,Anaelia Ovalle,Adina Williams,Levent Sagun
类目: Artificial Intelligence (cs.AI); Computers and Society (cs.CY)
备注: Accepted at Interspeech 2026
Abstract:In this paper, we motivate the need for a participatory speech dataset creation framework through a case study of the LGBTQIA+, or queer, community - a community with documented concerns about AI and reported harms, including attempts to develop ‘gaydar’ technologies that purportedly identify individuals as queer. We review common speech data collection practices, why these methods may be unsuitable for engaging with queer speakers, and discuss previous efforts in participatory AI with queer community engagement, as well as participatory endeavours specific to speech data collection for other marginalized communities. From this review, we develop a conceptual framework for participatory speech data curation by, for, and with marginalized communities drawing on insights from co-design and knowledge sharing. We propose a framework comprising overlapping and two-way processes of defining a community, project formulation, modes of participation, and personal autonomy.
[AI-122] Queer inclusion in speech datasets: An audit and taxonomy of practical tensions INTERSPEECH2026
链接: https://arxiv.org/abs/2609.25491
作者: Brooklyn Sheppard,Anaelia Ovalle,Adina Williams,Levent Sagun
类目: Artificial Intelligence (cs.AI)
备注: Accepted at Interspeech 2026
Abstract:In this paper, we examine speech datasets for their inclusion of LGBTQIA+, or queer, voices and provide a taxonomy of tensions to better understand why there is a lack of such voices in current speech technology datasets. Through an audit of six diverse speech datasets, we find that measurable queer representation is low (0-1.4% of speakers) - insufficient for robust disparity measurement. We take this community as a case study to consider what challenges and tensions are associated with collecting speech data from marginalized communities. For comparison, we audit an additional two datasets from the speech sciences that were created by, for, and with the queer community. We note that many customs in speech dataset collection efforts in AI and speech technology research may conflict with values emphasized in participatory approaches with marginalized communities, and provide a taxonomy describing these tensions.
[AI-123] rminal Shrinkage Averag ing Reveals a Schedule-Estimator Interaction in LLM Pretraining
链接: https://arxiv.org/abs/2609.25482
作者: Adam Ousherovitch,Yixin Wang
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
备注:
Abstract:Large language model (LLM) pretraining conventionally returns the raw final iterate. This couples two design choices: the learning-rate schedule that generates the parameter trajectory and the estimator that constructs the deployed model (e.g. the raw final iterate or a checkpoint average). A schedule that promotes optimization progress may differ from one that minimizes variation in the raw final iterate. Separating these choices creates an opportunity to maintain progress late in training while reducing variation in the returned model. To this end, we propose \emphTerminal Shrinkage Averaging (TSA), which interpolates between the raw final iterate and the average of recent checkpoints to balance recent progress against terminal variation. We analyze how TSA changes the preferred terminal learning-rate schedule under a local quadratic approximation and test this interaction through a sequence of controlled NanoChat experiments. Finally, we demonstrate that the resulting gains transfer to depth-22 NanoChat, where the combined schedule and estimator improve validation quality. A qualifying time-to-GPT-2 run also finishes faster than the public baseline used in our experiments, providing preliminary evidence of benchmark acceleration.
[AI-124] Spectra: A Rules-Driven LLM Pipeline for Automated KYC Document Processing
链接: https://arxiv.org/abs/2609.25474
作者: Miray Wahib,Ethan Tran,Rea Mourad,Mira Muti,Nikita Dvornik
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Know Your Client (KYC) onboarding in capital markets requires analysts to manually classify documents, extract structured data from heterogeneous sources, and validate compliance against complex regulatory policies. This process requires significant analyst time per client, with end-to-end onboarding often stretching to multiple weeks due to sequential handoffs. In this work, we analyze an on-boarding process and find that it comprises repeatable components well-suited to AI automation. We therefore propose a restructured workflow to be amenable to automation: we consolidate the traditional four-party process into two parties that share most of the work and can be automated together, eliminating intermediate handoffs that compound delays. To automate the remaining steps, we introduce Spectra, an AI-assisted document processing platform that combines a structured rules engine with LLM-based classification, extraction, and validation agents. The rules engine encodes compliance policy as a queryable database, enabling focused context injection that reduces token usage while improving extraction precision. Rather than a single monolithic prompt, the system decomposes document processing into isolated, auditable stages, each optimized independently and traceable to specific policy clauses. In evaluation on real KYC documents, Spectra achieves 100% classification accuracy and 89.4% extraction accuracy. Human review burden dropped by 96%.
[AI-125] A Practical Recipe for Semi-Supervised Federated ASR: Online Pseudo-Labels with Server Update Stabilization
链接: https://arxiv.org/abs/2609.25471
作者: Wonho Bae,Zakaria Aldeneh,Martin Pelikan,Jan “Honza” Silovsky,Tatiana Likhomanenko,Sheikh Shams Azam
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Machine Learning (stat.ML)
备注:
Abstract:Semi-supervised federated learning (SSFL) trains models on clients’ unlabeled data using a teacher to generate pseudo-labels, with a small labeled seed dataset on the server. Automatic Speech Recognition (ASR) is particularly fragile here: pseudo-label errors compound across the output sequence and across training rounds into divergence, leaving a large gap to fully-supervised FL. We show that closing this gap turns on two coupled design axes – the teacher (which model generates the pseudo-labels) and the anchor (the server-side updates on labeled data that stabilize training). On the teacher axis, a per-client online teacher (each client’s own evolving model) diverges on its own, but once stabilized it matches or beats the broadcast global teacher (one server model, fixed within a round) – decisively in-domain and competitively under domain shift. As the seed grows stronger and the online teacher’s advantage narrows, a transitioning teacher (global \rightarrow online at round r ) matches or beats both. On the anchor axis, the server must keep training on labeled data between rounds – otherwise the online teacher drifts – and this interleaving, more than the seed model, governs convergence. The two axes are inseparable: aggressive teacher choices pay off only once the anchor stabilizes training, which is highly sensitive to data augmentation and batch size – the settings that govern how much input and gradient noise the server injects. How much stabilization is needed is domain-dependent, governed by the dispersion of the seed data and its overlap with client data. These findings yield guidelines for SSFL in ASR training, improving over the strongest prior method on 9 of 11 pairs, by 20.8% on average in-domain and 10.0% cross-domain, narrowing the gap to fully-supervised FL.
[AI-126] RAG -NAROK: Retrieval-Aware Knowledge Corpus Poisoning in RAG with Source-specific Refutation
链接: https://arxiv.org/abs/2609.25469
作者: Abdullahil Kafi,Alvi Ataur Khalil
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Retrieval augmented generation (RAG) systems have emerged as the dominant architecture for grounding large language model (LLM) outputs in verifiable external knowledge, yet their structural reliance on a dynamic retrieval pipeline introduces a largely unexplored class of adversarial vulnerability. Existing knowledge-base poisoning attacks are fundamentally static. Adversarial documents are pre-computed and injected without any awareness of what the victim system will actually retrieve for a given query, leaving the attack blind to the competitive documentary landscape that surrounds its payload in the generator’s context window. Unlike traditional static poisoning attacks that are blind to the retrieved context, we introduce RAG-NAROK (Retrieval-Anchored Generation Negation And Response Quality Collapse), a RAG attack framework that adapts to the query text. RAG-NAROK exploits the transparency inherent in RAG pipeline to first extract the legitimate source identities, then generate Anchor-Specific Refutation documents that explicitly name and devalue retrieved sources while leveraging recency and authority biases to steer the text generation toward a target answer. Our results demonstrate that RAG-NAROK significantly outperforms static baselines across diverse domains, revealing a fundamental tension between RAG transparency and AI security.
[AI-127] ShowTellArena: Evaluating Business Workflow Understanding from Demonstrations
链接: https://arxiv.org/abs/2609.25467
作者: David Garg,Ritobrata Sarkar,Ehsan Azarnasab,Siddhartha Borah
类目: Artificial Intelligence (cs.AI)
备注: 6 pages, 1 figure, 2 tables. Code and dataset linked in the paper
Abstract:We often teach a colleague by showing the work and explaining the decisions as we go. How can we check what an agent understood from the same lesson? We introduce ShowTellArena, a benchmark protocol and public dataset for comprehension after narrated business demonstrations. The v1.0 release contains 50 business workflow tasks, with recordings, screenshots, narration, fixture seeds, and 502 questions. Tasks span finance, hiring, procurement, customer decisions, inventory, and logistics. The protocol holds the business scenario and quiz fixed while allowing each product to capture the lesson through its own teaching interface. Questions test operational rules, boundaries, exceptions, and errors in proposed automations. We analyze 218 selected pilot attempts across 39 workflow cases, including 28 cases attempted by all three evaluated systems. These exploratory results expose both answer errors and failures to complete the teaching experience. We describe the release’s verification gaps and the pilot’s uneven coverage, exclusions, and grading provenance. The contribution is an inspectable dataset and assessment workflow that others can extend; the selected pilot is not a controlled product ranking.
[AI-128] Real-Time Hand Gesture Recognition for OpenXR Using Transformer-Based Machine Learning
链接: https://arxiv.org/abs/2609.25466
作者: Salar Rezayani,Russell Butler
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Hand gesture recognition is a key component in human-computer interaction (HCI), enabling intuitive interfaces for applications in gaming, virtual reality (VR), robotics, and more. This study integrates transformer-based machine-learning models for real-time hand gesture recognition, using hand-tracking data captured through the OpenXR standard in Unity. We leverage positional data of hand joints and wrist rotation angles to train a custom gesture recognition system. By utilizing the sequential modeling capabilities of transformers, the system captures temporal dependencies within short gesture windows and classifies gestures robustly across hand orientations and sizes. The results show a significant improvement in gesture classification accuracy. Building on this, we outline how the approach can be extended toward detecting the flow of movement, i.e., the transitions between gestures, as future work.
[AI-129] Rollout Efficiency in Reinforcement Learning for Reasoning Large Language Models : A Taxonomy and Future Directions
链接: https://arxiv.org/abs/2609.25463
作者: Niloofar Gholipour,Marcos Assuncao,Gursimran Singh,Timothy Yu,Rajkumar Buyya,Julien Gascon-Samson,Zhenan Fan,Yong Zhang,Xiaojie Xu,Yaqiang Yao,Xiaolong Bai
类目: Artificial Intelligence (cs.AI); Distributed, Parallel, and Cluster Computing (cs.DC)
备注:
Abstract:Reasoning-oriented reinforcement learning enables large language models to solve mathematical, coding, and other multi-step tasks, but shifts a substantial portion of the training cost to rollout, where trajectories are generated for policy updates. Efficient rollout mechanisms are therefore essential to reduce this cost while maintaining the freshness, consistency, and statistical validity of training data. This survey provides a systematic taxonomy of recent research on rollout efficiency for reasoning-oriented reinforcement learning, classifying existing approaches from both mechanism and bottleneck perspectives. Based on this taxonomy, we analyze how different technique families address distinct sources of rollout inefficiency, examine opportunities and potential conflicts for combining them, identify gaps in the evaluation and reporting of efficiency gains, and discuss open challenges and future research directions.
[AI-130] ZeroGate: Trust-Preserving Fast Paths for Governed AI Agent Runtimes
链接: https://arxiv.org/abs/2609.25443
作者: Zexun Wang
类目: Artificial Intelligence (cs.AI); Cryptography and Security (cs.CR); Distributed, Parallel, and Cluster Computing (cs.DC)
备注: 47 pages, 11 figures, 13 tables. Full research report with an ancillary archive containing reference code, raw experimental records, and offline verification tools
Abstract:Moving authorization earlier can shorten an agent’s dispatch boundary without removing authorization work. It can also admit an action whose payload, authority, or relevant state has changed. ZeroGate separates exact-action approval from durable local admission: an issuer signs a short-lived ActionPass, and a trusted runtime adapter reconstructs the final action before a local gate checks its binding and consumes its nonce. A SQLite transaction couples nonce consumption, applicable quota updates, and an admission receipt. We state a conditional decision-preservation proposition: successful local admission implies that a specified synchronous policy would authorize the same action at the admission point, provided approval is sound, all policy dependencies are represented and current, observations are faithful, and consumption is atomic. The implementation alone establishes neither current-world freshness nor exactly-once remote effects. Evaluation separates authored semantic fixtures, controlled concurrency and crash experiments, and an Azure Blob study comparing synchronous and prepared execution through the same issuer and gate. Both modes mint an exact-action pass; lifecycle latency includes preparation and prepared-batch dwell. Across 4800 cloud attempts, prepared worker-admission-to-dispatch p95 ranges from 9.802 to 11.374 ms, versus 25.018 to 334.000 ms synchronously, across the tested concurrency levels. Prepared mean complete lifecycle is longer at every level: the boundary improvement is not a net speedup. The contribution is an explicit revalidation contract, a durable reference boundary, and an auditable comparison of where authorization cost is paid, not a new cryptographic primitive or a universal performance frontier.
[AI-131] Lightweight Ranking Heads: Accelerating Multi-Task Experimentation in Production Recommender Systems RECSYS2026
链接: https://arxiv.org/abs/2609.25433
作者: Sanjay Surendranath Girija,Aniruddh Nath,Li Wei,Yanhao Jiang,Shawn Andrews,Lukasz Heldt,Yi Wu,Aditya Mahajan,Mohit Sharma
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
备注: ACM RecSys 2026 - Online and Adaptive Recommender Systems
Abstract:Modern production-scale recommender systems rely on complex, multi-task ranking models. Introducing new prediction tasks into these massive systems often causes bottlenecks - it risks negative task conflicts with existing tasks, and can lead to long development and experimentation cycles due to the expensive retraining of backbone models and downstream models or tuning of reward combination formulas. To address the critical challenge of slow experimentation velocity, we introduce the Lightweight Ranking Heads (Light Heads) framework. Designed for continuous online learning environments, Light Heads enable the dynamic injection of new tasks into existing multi-task ranking models, effectively obviating the need for model cold-starting and retraining of backbone models. By utilizing stop-gradients and stateless daily training, this design strictly isolates new tasks, mitigating the risk of adverse task conflicts. Crucially, this framework uses a centralized configuration that allows Light Heads to be added to multiple models simultaneously, unblocking faster training data generation and co-training of downstream models. Successfully deployed at YouTube scale, this approach reduces the iteration cycle for multi-task experimentation from several weeks to days. In this paper, we detail the system architecture, analyze the training dynamics of stateless cold-started heads, compare their performance to full heads, and demonstrate how Light Heads have enabled the rapid A/B experimentation and deployment of new ranking tasks that yield measurable production value.
[AI-132] Predictive Uncertainty for Neural CAE Surrogates
链接: https://arxiv.org/abs/2609.25430
作者: Kaustubh Tangsali,Mohammad Amin Nabian,Kelvin Lee,Carmelo Gonzales,Sanjay Choudhry
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Probability (math.PR); Computational Physics (physics.comp-ph)
备注:
Abstract:Neural surrogates can substantially accelerate computer-aided engineering (CAE) workflows, but their use in design requires uncertainty estimates that remain meaningful across varying geometries, spatial prediction fields, and engineering quantities of interest. We investigate how established uncertainty quantification (UQ) approaches behave when adapted to geometry-conditioned neural surrogates. We compare one closed-form and two sampling-based approaches-a Gaussian process (GP)-based method, concrete Monte Carlo (MC) dropout, and deep ensembles-and evaluate them on three large, industry-relevant CAE datasets for external aerodynamics and crash dynamics. We examine whether predicted uncertainties have credible magnitudes, identify locations with larger prediction errors, respond to unfamiliar inputs, and remain informative for derived engineering quantities. On the DrivAerStar dataset, where all three methods are compared, each generally assigns higher uncertainty to locations with larger prediction errors, and validation-based rescaling brings interval coverage close to nominal on a disjoint in-distribution test set. Results on AirFRANS and automotive crash also show useful error ranking and interval estimates, but the relative performance of the methods changes with the dataset and evaluation criterion. UQ methods and evaluation metrics should therefore be selected based on the intended downstream CAE decision. Subjects: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Probability (math.PR); Computational Physics (physics.comp-ph) Cite as: arXiv:2609.25430 [cs.LG] (or arXiv:2609.25430v1 [cs.LG] for this version) https://doi.org/10.48550/arXiv.2609.25430 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
[AI-133] Beyond Natural Language: An Agent -Native Language for Autonomous Science
链接: https://arxiv.org/abs/2609.25421
作者: Yifeng He,Jiachen Liu
类目: Programming Languages (cs.PL); Artificial Intelligence (cs.AI)
备注:
Abstract:As autonomous AI agents take on every stage of scientific inquiry, research output is expanding far beyond human review capacity. Yet scientific communication still relies on natural-language prose: an informal medium prone to ambiguity, hidden assumptions, and untracked limitations that machines cannot reliably audit. We introduce Lara, a machine-checkable language and protocol for checking and revising support for research claims. By turning research arguments into executable artifacts, Lara provides an epistemic kernel for autonomous science: it enables automated validation pipelines for research agents, lets declared bridges connect arguments across papers into an auditable network, and allows both humans and machines to recheck the standing of an encoded claim in milliseconds. In a Lara program, authors explicitly declare their claims, supporting evidence and assumptions, and known objections or limitations. A lightweight, deterministic checker adjudicates these interactions, assigning each claim a reproducible status: “justified”, “defeated”, “contested”, or “gap”, which marks a claim whose support is incomplete and locates the unanswered question. Case studies cover empirical review, a philosophical debate without measurements, and the loss of support when an assumed axiom is withdrawn. We establish the metatheory of claim checking and cross-context argument transport, and mechanize the semantic guarantees in Lean 4 (roughly 117,000 lines), leaving three arguments on paper. The audited public metatheory is “sorry”-free and uses only Lean’s three standard axioms; some executable examples additionally trust native evaluation.
[AI-134] Robust Failure Conservative Repair: Textual Knowledge Distillation from Cross-Model Failures EMNLP2026
链接: https://arxiv.org/abs/2609.25400
作者: Andrew Ren,Haokun Liu,Chenhao Tan
类目: Artificial Intelligence (cs.AI)
备注: 16 pages, 2 figures. Accepted to EMNLP 2026 (Main track)
Abstract:Failure-based textual knowledge distillation aims to discover gaps in a model’s knowledge by examining its task errors. The distilled knowledge can be useful for the reasoning of both this model (“source model”) and other models. However, this transfer of knowledge may not be stable. We define a rule atom to be a standalone rule injected into a model’s textual input at inference time. A rule atom can encode transferable task knowledge or model-specific reasoning patches that can confuse other models. Also, the injected rule atoms can be misapplied to unrelated cases, causing the model to incorrectly flip its answer based on irrelevant information. Building on a pipeline that distills training examples into task-specific cheat sheets that aid model reasoning, we examine when failure-derived rules can improve these cheat sheets. Our early experiment shows rule distillation from a single model’s failures underperforms the baseline cheat sheet on non-source model families. This motivates Robust Failure, Conservative Repair (RFCR), a textual distillation procedure that derives rules from failures shared across models, sharpens their application boundaries using boundary cases, and abstains when no useful rule is found. On a 400-item BIG-Bench Hard task set, RFCR improves the baseline cheat sheets from 68.50% to 71.25% (+2.75 pp; 95% CI [+1.25,+4.50]) without performance degradation on previously correct cases. Ablations and cross-model diagnostics support that accuracy gains come from both new knowledge injection and strict rule-application control.
[AI-135] Deep Reinforcement Learning on Item-Compatibility Graphs for One-Dimensional Bin Packing
链接: https://arxiv.org/abs/2609.25397
作者: M. Aslı Aydın
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Optimization and Control (math.OC)
备注: 29 pages,3 figures, 8 tables
Abstract:The one-dimensional bin packing problem (1D-BPP) is a classical NP-hard combinatorial optimization problem with applications ranging from logistics and manufacturing to cloud resource management. Although deep reinforcement learning (DRL) has become a competitive paradigm for data-driven optimization, most learned packing methods target 2D and 3D variants, and intelligent learned solvers for 1D-BPP remain scarce. In this paper, we present a novel end-to-end, size-agnostic graph reinforcement learning framework for 1D-BPP. We formulate the packing process as a Markov decision process on an item-compatibility graph, serving as a structural knowledge representation in which every action merges two partial bins that fit together. A graph neural network actor-critic policy extracts relational features from this representation and is trained through reinforcement learning and decoded by stochastic beam search, enabling a single trained model to generalize zero-shot to instances of any size. We conduct a systematic empirical study across graph encoders, DRL algorithms, reward functions, training distributions, and hyperparameters. Evaluated zero-shot on the full BPPLIB benchmark against a constructive heuristic, a grouping genetic algorithm, and recent learned methods, our data-driven policy lowers the mean optimality gap of the constructive heuristic from 2.66% to 2.31%, with the largest gains on structured instances. Against learned baselines evaluated on the same benchmark, it attains a lower gap on most of the nine families and is far more stable across instance distributions. On the hardest benchmark family, it outperforms a state-of-the-art learned solver that relies on column generation and integer programming, while using no solver at all. A grouping genetic algorithm remains ahead overall, and we analyze where and why the residual gap arises.
[AI-136] VLAQuantBench: Closed-Loop Evaluation of Post-Training Quantization for Vision-Language-Action Models
链接: https://arxiv.org/abs/2609.25376
作者: Jiuyi Xu,Qing Jin,Meida Chen,Song Wang,Yang Sui,Yangming Shi
类目: Robotics (cs.RO); Artificial Intelligence (cs.AI)
备注: 28 pages, 35 tables, 4 figures
Abstract:Post-training quantization reduces the memory requirements of vision-language-action (VLA) models, but precision selection must account for the interaction between layer scope, numerical format, and calibration. We introduce \textbfVLAQuantBench, a controlled evaluation with 409 runs and 94,574 simulation episodes: four models on LIBERO, with X-VLA additionally evaluated on three simulation benchmark families. Under uncalibrated W4A4 round-to-nearest quantization, expanding a \pi_0.5 action-head subset from 126 to 167 layers raises success from 7.0% to 70.5%. Fixed-observation replay confirms a corresponding numerical recovery. Two-episode calibration removes the severe joint failures in the tested subsets, whereas the same smoothing-and-clipping recipe lowers \pi_0 success and does not recover OpenVLA-OFT end-to-end. For OpenVLA-OFT, protecting one 28,672-parameter output projection instead restores near-baseline success: the remaining 441 eligible linear layers retain W3 on LIBERO-Long or eight-bit activations across all four suites. Task-clustered intervals support the large failure and recovery contrasts. These results establish recipe-dependent interactions and identify concrete precision assignments, rather than universal layer-sensitivity rules. Real-kernel and physical-robot measurements complement the accuracy analysis. Code, configurations, and episode records are publicly available at this https URL.
[AI-137] From Decorative to Load-Bearing: Task Difficulty Shapes the Causal Role of Chain-of-Thought
链接: https://arxiv.org/abs/2609.25366
作者: Renee Jia,Di Mu
类目: Artificial Intelligence (cs.AI)
备注: Accepted to Transactions on Machine Learning Research (TMLR), September 2026. Code/ dataset available at the project repository and huggingface
Abstract:Chain-of-thought (CoT) monitoring is only meaningful if written reasoning causally constrains the answer. We introduce continuation-based causal testing, an ablation-patch intervention that perturbs one reasoning step, truncates the chain, and forces the model to continue from the corrupted prefix. It measures how load-bearing a CoT is for the final answer, a behavioral notion distinct from mechanistic faithfulness. Across Gemma-2-9B-IT, Llama-3.1-8B-Instruct, and DeepSeek-R1-Distill-Qwen-7B on GSM8K, MMLU, and BIG-Bench Hard, CoT load-bearingness tracks model-relative task difficulty: on easy tasks models silently bypass their own reasoning; on hard tasks they follow corrupted steps and propagate errors. A matched 2x2 analysis shows task difficulty dominates perturbation type: error propagation rises 16x from GSM8K to BBH multistep arithmetic, and a variance partition over 28,584 continuations attributes 98.8% of explained deviance to task difficulty versus 0.8% to perturbation type. Reasoning-specific RL suppresses error propagation and compresses the gradient. A four-variant judge-sensitivity analysis and blind two-annotator study (n=500) show the error-propagation vs. non-propagation label is invariant to judge prompt, with perfect inter-annotator agreement (Cohen’s kappa = 1.00). This gradient creates a structural problem for CoT-based oversight and AI safety monitoring: where the trace is easy to read it carries little signal, and where it matters errors propagate before a monitor can intervene. Linear probes on hidden states separate silent bypass, self-correction, and error propagation, but additive activation steering provides limited causal control, flipping only about 25% of error-propagation cases at best. Behavioral mode is readable but not reliably controllable.
[AI-138] Clarification Is Not Correction: LLM s Fail to Let Go
链接: https://arxiv.org/abs/2609.25337
作者: Jianzhe Lin,Xiaolin Li,Fei Wang,Robert Douglas,Rajeshkumar Golani,Jubin Chheda
类目: Artificial Intelligence (cs.AI)
备注: 17 pages
Abstract:Dialogue failures in language models are usually framed as memory failures: context too long, summaries lossy, a constraint forgotten. We argue this misses a deeper problem: in many conversations the model does not forget, it commits too early. An ambiguous early turn collapses into a single hidden interpretation, and later clarification is filtered through that commitment. We call this early posterior collapse: unresolved user intent collapsing into a committed task state before ambiguity is resolved. We study it with controlled dialogue tasks in writing, planning, and coding using Gemini-2.5-Pro and Gemini-2.5-Flash. Across thousands of trials, the same information in different orders yields different outcomes, even when the final dialogue contains equivalent task-relevant information. This order effect suggests later clarification is treated as extra context rather than a corrective signal: it refines a stale task state without invalidating it. Coding tasks are especially vulnerable, suggesting early assumptions get embedded in structured artifacts such as interfaces and control flow. Standard prompting and memory strategies do not reliably help: summaries can collapse ambiguity, and chain-of-thought can reduce explicit wrong commitment in reasoning traces without improving final task success. These findings motivate uncertainty-preserving state management. If assistants cannot let go of early interpretations, robustness cannot rely on post hoc correction alone; it must keep ambiguous early turns from hardening into one task state. Assistants should hold tentative hypotheses while ambiguity remains, ask before executing when high-impact ambiguity persists, and rebuild from a revised state when later evidence invalidates an earlier reading. Rather than one prompting fix, we aim to redirect research for interactive LLMs from retaining more context toward preserving uncertainty.
[AI-139] Potential for Enhanced Learning in Machine Learning Classes by Using Wiki LLM Indexing
链接: https://arxiv.org/abs/2609.25303
作者: Brian Wright
类目: Artificial Intelligence (cs.AI); Computation (stat.CO)
备注:
Abstract:Large language models are increasingly deployed as course-specific tutors, but their usefulness depends on grounding in vetted instructional materials that are often revised mid-semester. Our prior work built a multimodal retrieval-augmented generation (RAG) system over an authentic machine learning course corpus (Foundations of Machine Learning) and found that retrieval improved contextual grounding, but that fixed retrieval strategies were suboptimal. That motivates a different question: whether how a corpus is structured at ingest time matters more than how much is retrieved at query time. We present a controlled head-to-head comparison of two knowledge representations over an identical classroom corpus: (A) vector RAG, replicating the best-performing configuration from our prior study, and (B) an LLM-compiled wiki (Karpathy framework), in which the corpus is synthesized at ingest into linked concept pages with explicit cross-references and citations back to source materials. We evaluate 59 questions spanning single-fact recall, cross-unit concept linking, synthesis and explanation, and currency after a syllabus revision, scored by an LLM judge against a human-authored rubric. Both representations answered single-fact questions about equally well (9.33 vs. 9.96 of 10), but diverged sharply on questions requiring links across course units. The compiled wiki remained accurate and grounded (9.93; 100% grounded in cited sources), while retrieval scored lower and was markedly less grounded (8.14; 64%). The wiki’s citations let students and instructors trace any claim back to the lecture that introduced it, adding a layer of dynamic retrieval that machine learning courses require. While further testing is needed, instructors using AI to support learning in ML courses should consider wiki-based structure for its potential to support foundational elements of best practice.
[AI-140] Making Agents More Consistent: Skills Should Form Habits for Repeat Tasks
链接: https://arxiv.org/abs/2609.25299
作者: Travis Weber,Rohit Taneja
类目: Artificial Intelligence (cs.AI); Cryptography and Security (cs.CR)
备注: 12 pages, 2 figures
Abstract:On repeated work, agents are inconsistent. We ran 42 tasks three times each and found that, depending on the model, 38% to 74% returned answers that did not agree. Consistency is what a buyer, an auditor, or a regulator requires, and agents do not have it. They are wasteful too: 95.3% to 97.2% of what an agent generates goes to re-deriving a plan the system already knows. We propose skill habit formation. An agent mines its own execution history for candidate skills, deterministic variants that compete against the incumbent rather than replacing it. A candidate declares the region of input space it claims, so the common case runs as a script and the rest falls through to reasoning. Four gates of ascending cost admit candidates; the central one tests a candidate’s execution trace against a retained reference, within a tolerance measured from that reference’s own run-to-run variability. On text-to-SQL, three of four reasoning arms reproduced their own output on 11 to 13 of 42 repeated questions and the fourth on 26 of 42, while a habit-formed variant reproduced on all 456 dispatches we repeated and was non-inferior to every arm it replaced (p0.0001). It also used 14% to 56% fewer tokens, turning net positive after 7 to 53 reuses. We measured what this costs in accuracy. The guard admitted work it should have deferred on 2.6% of natural paraphrases and 26% of inputs near its boundary, and 11 of 13 such failures were invisible to the trace-conformance gate at any threshold. Deterministic errors repeat exactly: a bad habit is as reliable as a good one, and that is the price of the property that makes the system auditable. Separating routing from parameter extraction raised end-to-end accuracy from 0.888 to 0.952 at 43% of the cost. Comments: 12 pages, 2 figures Subjects: Artificial Intelligence (cs.AI); Cryptography and Security (cs.CR) ACMclasses: I.2.11; D.2.5; D.2.4 Cite as: arXiv:2609.25299 [cs.AI] (or arXiv:2609.25299v1 [cs.AI] for this version) https://doi.org/10.48550/arXiv.2609.25299 Focus to learn more arXiv-issued DOI via DataCite (pending registration) Submission history From: Rohit Taneja [view email] [v1] Mon, 21 Sep 2026 18:43:49 UTC (30 KB)
[AI-141] Learned Enterprise Data Comprehension: Compression and Routing for Data Agents
链接: https://arxiv.org/abs/2609.25286
作者: Ethan Torres,Eric Mills
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Structured-data agents in enterprise settings must reason over complex data environments whose relevant evidence is distributed across schemas, relationships, policies, and recurring business roles. Modern agentic systems often address this burden through reusable markdown-style memory or skill files that preserve previously discovered information for later queries, reducing the need to rediscover the same structure repeatedly. This is useful, but it obscures a natural division of labor: agents are well suited to semantic reasoning, while learned systems are well suited to predicting and organizing recurring structure. We introduce latent equivalence learning to bridge this gap. The framework separates persistent task-relevant identities from their dataset-relative realizations. In our realization, supporting and opposing evidence shape support-realized Gaussian prototypes that learn how those identities are expressed in a particular data environment, while soft-membership profiles retain distinctions lost under a hard assignment. A separate learned query-prototype system represents recurring evidential requirements and maps them through a learned compatibility function into the same persistent identity structure. This identity-factorized, query-conditioned routing materializes the relevant dataset-specific evidence for downstream reasoning, allowing the agent to operate over an already organized evidential state rather than reconstructing cross-schema structure at every query. On the Data Agent Benchmark, spanning 54 queries across 12 heterogeneous datasets, our full implementation achieves 94.67% dataset-macro stratified Pass@1 over five complete trials and 258/270 successful raw query attempts, compared with 55.51% for the benchmark’s Claude Opus 4.6 reference agent, ranking first among 40 leaderboard entries at submission.
[AI-142] Attention as a Routing Graph: Live Circuit Extraction from a Single Forward Pass
链接: https://arxiv.org/abs/2609.25285
作者: Ash Manvi,Samreena Tajreen
类目: Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
备注: 7 Pages, 5 Figures, 1 table, Code: this https URL
Abstract:Finding circuits in language models usually means running many careful interventions. We try something simpler: treat attention as a routing map from one forward pass, keep a small set of routes that point toward the answer, and ask whether those routes actually matter. They often do. On induction and IOI (tasks where the “right” circuit is already known), ablating our extracted edges hurts the model much more than ablating a random set of the same size. We evaluate n=100 prompts per cell on GPT-2 Small, GPT-2 Medium, and Pythia-410M, with paired gap tests and bootstrap confidence intervals. The extract step costs one forward; a head-by-head patch sweep costs about two orders of magnitude more. We are not claiming a complete circuit atlas. We are claiming a cheap sketch that carries real causal signal on known tasks, with clear failure modes when it does not. Code and evaluation artifacts are at this https URL. Comments: 7 Pages, 5 Figures, 1 table, Code: this https URL Subjects: Artificial Intelligence (cs.AI); Machine Learning (cs.LG) Cite as: arXiv:2609.25285 [cs.AI] (or arXiv:2609.25285v1 [cs.AI] for this version) https://doi.org/10.48550/arXiv.2609.25285 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
[AI-143] When LLM Agents Fail to Read the Room: ReAdapt for Relational Social Reasoning
链接: https://arxiv.org/abs/2609.25284
作者: Jianzhe Lin,Xiaolin Li,Yunda Liu,Fei Wang,Jubin Chheda
类目: Artificial Intelligence (cs.AI)
备注: 12 pages
Abstract:A social agent’s most basic decisions (should I react to this post? who should I reach out to?) are not purely content problems. The right action often hinges on the latent relationship between people – tie strength, reciprocity, mutual connections – rather than on which content is most salient. Standard LLM agent loops do not explicitly represent how new relational evidence should revise the agent’s current social hypothesis, leaving them prone to surface-obvious choices when relational and content cues diverge. We formalize this failure mode with a relationship-reasoning benchmark: 500 synthetic social worlds with friendships, follows, reaction histories, and feeds, yielding 1,000 queries over two tasks, reaction selection and warm introduction (finding the best bridge to a target person). By construction, the surface-obvious candidate differs from the relationship-grounded oracle in about 53% of queries, forming an overturn subset where the agent must use relational evidence to revise an initially plausible choice. We propose ReAdapt (Relationship-Adaptive Agent with Policy-driven sTate), which augments the ReAct loop with an explicit structured social state z = (G, B, R, N, D) capturing goal, belief, relationship, norm, and disclosure. After each tool observation, ReAdapt runs a typed Adapt step that updates this state and emits a policy operation (continue, switch, abandon, or clarify) before choosing the next action. With Gemini-3-Flash on a stratified subset of n = 150 queries per task, ReAdapt improves warm-introduction accuracy from 37% to 51% (+14 points) and reaction-selection accuracy from 69% to 77% (+8 points). Oracle regret drops from 0.260 to 0.152 and from 0.095 to 0.053, respectively. Holding the model, tools, and environments fixed, these results suggest that explicit relational-state adaptation helps LLM agents turn retrieved social evidence into revised decisions.
[AI-144] MedGate-Fusion: Integrating First-Encounter Semantic Narratives and Physiological Biomarkers for Prospective Stroke Risk Stratification
链接: https://arxiv.org/abs/2609.25272
作者: Hemn Khdr,Mohammad Noaeen,Karim Keshavjee,Aziz Guergachi,Zahra Shakeri
类目: Artificial Intelligence (cs.AI)
备注: 7 pages,2 tables, 2 figures. Submitted to EMBC 2026
Abstract:Prospective stroke risk stratification in primary care is challenging because early risk signals are distributed across routine biomarkers and unstructured clinical narratives. We propose MedGate-Fusion, a multi-modal gated architecture that integrates transformer-based embeddings of first-encounter narratives with ten routinely recorded risk markers. We used electronic medical record data from the Canadian Primary Care Sentinel Surveillance Network (CPCSSN). Starting from 808,921 encounter-level observations, we constructed a first-encounter cohort and retained 102,736 unique patient records with non-empty narratives and sufficient data to evaluate a five-year stroke outcome. To reduce explicit target leakage from diagnostic mentions in notes, we applied dictionary-based redaction of stroke-related terms prior to semantic encoding.
[AI-145] he AI Neuroscientist: An Interactive Agent ic Interface for Neuroimaging Analysis
链接: https://arxiv.org/abs/2609.25254
作者: Aakash Patel,Panos Ketonis,Shreya Saxena,Smita Krishnaswamy,David van Dijk
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Analyzing neuroimaging data requires specialized coding and statistical expertise, which limits accessibility for researchers without computational backgrounds. We present the AI Neuroscientist, a language agent for interactive data exploration. The system integrates a large language model (LLM) with a neuroimaging toolset to perform quality control, modeling, and visualization. This allows researchers to query data quality and specify analysis parameters directly in natural language, providing a transparent and interactive alternative to conventional scripted pipelines for small-scale data exploration. We demonstrate these capabilities using functional near-infrared spectroscopy (fNIRS) data, and evaluate the agent on a custom fNIRS benchmarking suite against general-purpose LLM agents with code sandboxes. Future extensions will generalize the architecture to additional modalities, including functional magnetic resonance imaging (fMRI) data, and expand the benchmarking suite to additional fNIRS tasks.
[AI-146] Lean Pool: An AI-Maintained Archive of Formalized Mathematics
链接: https://arxiv.org/abs/2609.25199
作者: Vasily Ilin
类目: Artificial Intelligence (cs.AI)
备注: 52 pages, 6 figures. Includes a catalogue of imported projects
Abstract:Lean Pool is a repository of formalized mathematics. It is grown, maintained and optimized by AI agents.
[AI-147] X-Planner: Event-Structured Task Planning for Embodied Intelligence
链接: https://arxiv.org/abs/2609.25187
作者: Howard Lu,Shalfun Li,Porter Pan,Cris,Lumen,Cyril,Eric Hu,Lily Li,Maeve Zhang,Robert Wang,KZ Zheng,Viggo Chen,Tim Ding,Regsis Cheng,YJ Xiao,Kian,Hai Lin,Alan Song,Elise Ma,Gody Li,Victor Yao,Yohann Tang,Ingrid Yu,Jason He,James Wang,Ryan Yu,Ping Yang,Chris Pan,Vincent Chen,Roy Gan,Hao Wang,Qian Wang
类目: Artificial Intelligence (cs.AI)
备注: this https URL
Abstract:Task planning bridges high-level instructions and executable behavior in long-horizon manipulation, yet modern Vision-Language-Action (VLA) systems often leave this intermediate structure implicit. Existing chain-of-thought (CoT) planners also tend to rely on coarse task-level annotations or serialize long reasoning traces token by token. We present X-Planner, a planning front-end that addresses both the supervision and representation of embodied reasoning. Our planning data combine Ego, UMI, and teleoperation under a hierarchy granularity with source-dependent annotation depth. Takeover-time annotations and human-designed failures supervise ongoing error recognition. On the model side, a shared VLM backbone exposes two event-structured plan forms: a discrete interface that emits interpretable event states and a latent interface that relays continuous CoT states across staggered Transformer depths through Staircase Decoding. A frozen latent-to-text reconstruction objective provides a semantic anchor for the latent representation. Offline two-step planning evaluation places X-Planner second among four evaluated models on both BERTScore-F1 and a judge-based Overall score. In real-robot experiments, respectively, outperforming the evaluated baselines. These results characterize planning-text quality and downstream execution.
[AI-148] Multi-Term Fourier Graph Neural Network with Sample Relationship Learning for Enhanced Remaining Useful Life Prediction
链接: https://arxiv.org/abs/2609.25179
作者: Ya Song,Laurens Bliek,Yaoxin Wu,Yingqian Zhang
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
备注:
Abstract:Predicting the remaining useful life (RUL) is essential for effective predictive maintenance. Spatio-Temporal Graph Neural Networks (ST-GNNs), which can model both temporal and spatial relationships by representing time series data as a sequence of graphs, have shown exceptional performance in RUL prediction. However, current ST-GNNs face several drawbacks. First, they require domain expertise or significant computational power to establish graph structures prior to deploying GNNs. Second, the models are restricted to capture temporal dependencies within a predefined fixed-size lookback window. This restriction ignores the common issue of varying time series lengths, leading the prediction model to miss short-term or long-term dependencies. Finally, conventional models often fail to capture the inherent relationships between samples generated from adjacent time windows, which are crucial for improving both the accuracy and robustness of predictions. To address the aforementioned issues, we introduce a novel framework called Multi-Term Fourier Graph Neural Network with Sample Relationship Learning (MTFGN-SRL). Rather than treating the sample as a sequence of graphs, we consider it as a single complete graph and utilize a Fourier Graph Neural Network (FGN) to capture the spatio-temporal information in the frequency domain. We propose a multi-term learning module that utilizes multiple lookback windows to generate samples with varying terms, which are then fed into the FGN to enhance the extraction of useful information from the data. Finally, we develop a sample relationship learning module by training a heterogeneous GNN to identify inter-sample relationships, resulting in enhanced accuracy and robustness in predictions. Evaluations on the CMAPSS dataset demonstrate MTFGN-SRL’s superior performance over state-of-the-art methods in RUL prediction.
[AI-149] Mitigating Sequential Reappearance in Diffusion Data-Point Unlearning
链接: https://arxiv.org/abs/2609.25166
作者: Donghyun Kim,Taehyuk Lee,Jinyeong Kim,Youngmin Oh,Dohyeong Kim,Jaehyuk Ryu,Sangwoo Hong
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
备注:
Abstract:Diffusion data-point unlearning is typically evaluated immediately after each deletion, even though subsequent requests may repeatedly update the same model. We identify sequential reappearance, a failure mode in which an instance that is initially judged to be forgotten later returns to the memorized regime without reuse of the deleted data or adversarial fine-tuning. To capture this behavior, we introduce a target-level evaluation protocol that tracks whether each target is forgotten immediately, remains forgotten at the end of the sequence, or reappears during subsequent deletions. We further find that targets that later reappear exhibit sharper local denoising-loss geometry after deletion than targets that remain forgotten.
[AI-150] Ovis-Embedding: Pushing the Frontiers of Universal Omni-Modal Embeddings
链接: https://arxiv.org/abs/2609.25165
作者: Ovis-Embedding Team
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:In this report, we introduce \textbfOvis-Embedding, a state-of-the-art omni-modal embedding family built on native integration of text, image, video, and audio. Instead of assembling separate modality towers, Ovis-Embedding uses a shared multimodal backbone to encode different modalities in a common representation space. Specifically, we make \textbfthree key advances: (1) \textbfnative omni-modal initialization: we adopt a pretrained Qwen-omni model as the embedding backbone and adapt it through contrastive training with low-rank initialization; (2) \textbfdata-centric omni-modal training: we construct a broad, high-quality corpus spanning text, images, video, audio, and interleaved multimodal data. To improve data efficiency, we introduce homogeneous-source sampling to form task-consistent batches with informative in-batch negatives; and (3) \textbfembedding-specific training and inference optimization: we use focal loss to emphasize hard examples and similarity-based Embedding Distillation to transfer fine-grained similarity structure from complementary experts. At inference time, low-rank feature decomposition enables compact embeddings with flexible dimensionality and minimal performance loss. Empirical evaluations show that the \textbfOvis-Embedding family achieves state-of-the-art performance on \textbfMMEB-v3, \textbfMMEB-v2, \textbfMVEB, \textbfMAEB, and \textbfRTEB, demonstrating its effectiveness across text, image, video, and audio modalities. These results highlight the potential of unified omni-modal training to overcome modality fragmentation and advance universal embedding models for any-to-any retrieval.
[AI-151] Benchmarking Neural Defend ARCAS 1B: A Foundational Multimodal Deepfake Detection Model
链接: https://arxiv.org/abs/2609.25154
作者: Sivashankar Selvarajan,Piyush Verma,Sumit Kumar,Sharayu N. Deshmukh
类目: Cryptography and Security (cs.CR); Artificial Intelligence (cs.AI)
备注:
Abstract:AI-generated imagery evolves faster than benchmark-specific detector evaluations, making a single score an incomplete account of generalization. This paper evaluates Neural Defend ARCAS 1B across benchmark families without benchmark-specific parameter updates. We retain native aggregation and supplement it with record-level measures, coverage accounting, and subgroup diagnostics. Each Results subsection identifies the release and evaluation population, reports the official metric, and describes observed error patterns. A combined analysis synthesizes shared patterns while preserving the distinction between native and pooled quantities. Cross-paper comparisons are restricted to aligned evidence; differences in release, population, preprocessing, training, or benchmark exposure are context rather than rank. The findings characterize performance on evaluated records, not universal reliability, calibration, attribution, or future adaptive attacks. By keeping benchmark-native outcomes distinct from pooled summaries, the study makes test-population, class-balance, and missing-record-coverage differences visible. It supports interpretation of detector results in research, platform-safety, and forensic-review settings, foregrounding traceable protocol conditions over claims or leaderboard comparisons.
[AI-152] Exposing Blind Spots in Deep Imbalanced Regression Evaluation
链接: https://arxiv.org/abs/2609.25152
作者: Noah C. Puetz,Jens U. Brandt,Marc Hilbert,Elena Raponi,Thomas Bäck,Thomas Bartz-Beielstein
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
备注:
Abstract:Deep Imbalanced Regression (DIR) addresses a common failure mode of regression models: target distributions are highly non-uniform, causing models to perform best in densely populated target regions even when reliable performance is required across the full target range. Despite rapid methodological progress, DIR evaluation remains constrained by three blind spots: it is dominated by image-based benchmarks, its standard many-/medium-/few-shot protocol is diagnostic but not decision-complete, and tail-region stability across random seeds has not been systematically evaluated. We revisit DIR evaluation along these three axes. First, we broaden the data domain by evaluating DIR on a multimodal virtual sensing benchmark (\textscMuViS) with nine time-series extrinsic regression tasks across six physical domains, where rare target values often correspond to operationally meaningful regimes. Second, we adopt balanced MAE (\emphbMAE) and introduce balanced Mean Absolute Scaled Error (\emphbMASE), a scale-normalized metric for decision-complete comparison across methods and datasets. Third, through a repeated reevaluation of six representative DIR methods across multiple random seeds, we show that the tail regions targeted by DIR exhibit particularly high sensitivity to seed-level variability. Our results show that standard virtual-sensing models exhibit substantial tail degradation hidden by global MAE, that existing DIR methods can improve balanced performance but transfer unevenly to multimodal time-series data, and that tail-region instability remains a largely hidden failure mode under current DIR evaluation practice. Together, these findings and our publicly available code provide a reproducible basis for future DIR research toward regression systems that capture rare target regimes as reliably as common ones.
[AI-153] Brain-Inspired Hierarchical Modularity for General Continual Learning
链接: https://arxiv.org/abs/2609.25146
作者: Hongwei Yan,Kanglei Zhou,Qi Cheng,Weiyi Dong,Chunyan Lan,Guanglong Sun,Jun Zhou,Qian Li,Yi Zhong,Liyuan Wang
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
备注: 50 pages
Abstract:Continual learning, the ability to learn from sequential experience while retaining and adapting prior knowledge, is central to intelligent systems operating in changing environments. However, conventional continual learning is typically studied with offline task-wise training and clear task boundaries, leaving a substantial gap from general continual learning under online, uncertain, and evolving data streams. In this regime, intelligent systems must separate conflicting experience to reduce interference while integrating compatible experience to promote generalization. Inspired by the organization of the Drosophila learning and memory system, we identify a hierarchical modular principle that coordinates both functions through expert specialization and ensemble integration. We instantiate this principle as lightweight modular adaptation of pretrained foundation models, combining brain-inspired random expansion for expert routing and diversified modular integration across spatial and temporal scales. Across visual recognition, vision-language understanding, ego-exo video understanding, and embodied vision-language-action learning, our method consistently improves learning under online and uncertain data streams, with gains exceeding 50 percentage points over replay-free alternatives in embodied manipulation. These findings support hierarchical modularity as a biologically grounded path for learning from dynamic experience.
[AI-154] Stable Unsupervised Continual Chunking with Sheaf SyncMap
链接: https://arxiv.org/abs/2609.25143
作者: Xueyuan Li,Danilo Vasconcellos Vargas
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
备注:
Abstract:Unsupervised Continual chunking is a fundamental problem in machine learning and neuroscience, where the goal is to identify groups of states that frequently co-occur in temporal sequences. A key challenge is to form accurate chunks while maintaining their stability over time. In this work, we propose sheaf regularization to reduce local inconsistencies in Decentralized SyncMap, a self-organizing system, and thereby stabilize its chunking dynamics. We introduce a radial sheaf structure that penalizes distance-dependent radial motion between pairs of variables. Experimental results show that the proposed method achieves the highest normalized mutual information (NMI) among the evaluated SyncMap variants on 12 of 18 probabilistic Continual General Chunking Problem (CGCP) graphs with two-state memory and on 17 of 18 graphs with dynamic memory. In the sequential adaptation experiment, Sheaf SyncMap also achieves high NMI after shifts in the input distribution, indicating that it can adapt to new knowledge while avoiding the negative transfer commonly observed in modern machine learning systems such as neural networks.
[AI-155] he Probabilistic Structure of Large Language Models
链接: https://arxiv.org/abs/2609.25134
作者: Adnan Aboulalaâ
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Probability (math.PR); Machine Learning (stat.ML)
备注: Expository article on the probabilistic structure of large language models, 27 pages
Abstract:This paper presents a probabilistic perspective on large language models (LLMs), developed with the aim of bringing together, in a single self-contained account, tools that are usually treated separately across the literature. LLMs are described through probability measures on the set of sequences of tokens, specified via their autoregressive conditional distributions. Training is formulated as a maximum-likelihood estimation problem, addressed by stochastic gradient methods, while text generation is viewed as the sequential simulation of the resulting stochastic process. The role of the asymmetry of the Kullback–Leibler divergence in text generation is examined in relation with characteristic phenomena such as hallucination and the distinction between statistical plausibility and truth. As a complementary illustration of the same viewpoint, we also discuss diffusion models, built around the score function, which cast generation not as sequential token prediction but as the simulation of a reverse-time stochastic process transforming noise into data both in discrete and continuous time.
[AI-156] An Accurate and Interpretable Hyper Graph Neural Network for GBM Survival Prediction
链接: https://arxiv.org/abs/2609.25088
作者: Mushahid Intesum
类目: Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
备注:
Abstract:Survival prediction for glioblastoma multiforme (GBM) demands models that are both accurate and interpretable, yet existing approaches treat these objectives as com- peting, where performant models sacrifice transparency, while interpretable models accept degraded predictive power. We argue that this trade-off is not inherent. Graph neural net- works offer a structural foundation for extracting interpretable, explainable representations without compromising discriminative ability. Furthermore, current methods typically rely on a single imaging modality, underutilizing the complementary information available across multi-modal MRI and clinical metadata. We propose a multi-modal framework that inte- grates three components to address both objectives simultaneously: (1) a sheaf hypergraph neural network that captures higher-order relationships among tissue patches through direc- tional, asymmetric message passing; (2) a concept bottleneck layer that compresses learned representations into clinically grounded concepts, enforcing ante-hoc interpretability; and (3) an extension sufficiency test (EST) regularizer that penalizes unfaithful explanations during training, ensuring that model explanations genuinely reflect the internal decision process. Clinical and genomic features are incorporated through gated fusion, preserving the dominant prognostic signal of molecular markers while retaining concept-level traceabil- ity. Evaluated on 593 patients from the UPenn-GBM dataset under 5-fold cross-validation, our framework achieves a concordance index of 0.643 with the lowest fold-level variance among all compared models (std = 0.015). To our knowledge, this is the first work to unify sheaf hypergraph convolution, concept bottleneck supervision, and EST regularization for interpretable survival prediction from brain MRI
[AI-157] Federating Quantum and Classical Computing: A Privacy-Preserving Hybrid Approach
链接: https://arxiv.org/abs/2609.25082
作者: Carlos Cano,Daniel M. Jimenez-Gutierrez,Diego Sal,Georgios Kellaris,Joaquin del Rio,Oleksii Sliusarenko,Xabi Uribe-Etxebarria
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Distributed, Parallel, and Cluster Computing (cs.DC)
备注:
Abstract:Quantum machine learning (QML) is increasingly recognized as one of the most promising near-term applications of quantum computing, viewed as a next-frontier candidate beyond purely classical approaches. Hybrid quantum-classical models operationalize this potential by embedding a parameterized quantum circuit within a model where all other components remain classical-a design already applied to chemistry simulation, financial modeling, and image classification. However, their deployment in privacy-sensitive, multi-party settings is constrained by the need to avoid centralizing raw data and by the requirement that modern quantum circuits remain parameter-efficient to stay trainable at scale. In this paper, we address these constraints by evaluating federated learning (FL) as a means of combining a hybrid quantum-classical active party with a classical passive party, using this http URL’s Blind Vertical FL (SBVFL) protocol to avoid centralizing raw data, while drastically reducing communication. We construct the split multiplicative periodic parity (SMPP) benchmark, following common QML design practice. On this task, our simulations show that SBVFL raises accuracy from 0.7227 to 0.8757 compared to local training, closely approaching non-private centralized accuracy, and that the hybrid quantum-classical model achieves this with substantially fewer trainable parameters than the classical neural networks and random forest alternatives. These results show that FL enables high-performing, privacy-preserving quantum-classical collaboration without centralizing raw data. Subjects: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Distributed, Parallel, and Cluster Computing (cs.DC) Cite as: arXiv:2609.25082 [cs.LG] (or arXiv:2609.25082v1 [cs.LG] for this version) https://doi.org/10.48550/arXiv.2609.25082 Focus to learn more arXiv-issued DOI via DataCite
[AI-158] 4DGS-JEPA: Temporally Compositional Joint-Embedding Prediction for Dynamic Gaussian Splatting
链接: https://arxiv.org/abs/2609.25036
作者: Yongchao Huang
类目: Artificial Intelligence (cs.AI)
备注: 44 pages
Abstract:Dynamic Gaussian Splatting provides an explicit representation of evolving 3D scenes, but existing approaches are primarily optimized for reconstruction, future-state generation, or rendering rather than for learning reusable predictive dynamics. We propose 4DGS-JEPA, a Gaussian-native joint-embedding predictive architecture for causal multi-horizon prediction over dynamic Gaussian scenes. The model uses a hierarchical scene-, motion-group-, and Gaussian-level representation together with a horizon-conditioned transition operator that supports both direct prediction and recursive rollout. Its central principle is temporal composition: different chronological transition paths reaching the same future endpoint should produce compatible predictive states. Endpoint and multi-horizon path supervision anchor these predictions to future target embeddings, while a selective geometry decoder and geometry-level composition ground the learned dynamics in consistent group motion and Gaussian geometry without requiring complete future appearance reconstruction. We further introduce a hybrid correspondence mechanism that combines persistent canonical identity with residual optimal-transport matching under reordering and topology change. We characterize zero-loss path agreement and finite-error rollout accumulation theoretically. Three controlled experiments provide mechanism-level evidence that temporal composition reduces latent path dependence while retaining predictive accuracy, geometry-level composition improves consistency of decoded motion, and hybrid correspondence preserves reliable identity while remaining robust when correspondence becomes ambiguous. Together, 4DGS-JEPA provides a predictive, temporally compositional formulation of dynamic Gaussian worlds.
[AI-159] Not All 4-bit Quantizers Are Equal: Deployment-Time Mitigation of PII Leakage in Fine-Tuned Small Language Models
链接: https://arxiv.org/abs/2609.25014
作者: Cristhian Kapelinski,Diego Kreutz
类目: Cryptography and Security (cs.CR); Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
备注: 10 pages and 5 tables. The source code and per-seed experimental results are publicly available at this https URL . Accepted for publication at SBSeg 2026
Abstract:Organizations fine-tune small language models on private data and then compress them to 4 bits for resource-efficient deployment. We show that the compression method also affects privacy. What separates the methods is not the bit width but whether they tune their rounding on a small sample of text, the calibration corpus. On our primary model, when each planted record’s own opening text is used as the prompt, the two calibration-based methods we test, Activation-aware Weight Quantization (AWQ) and Gradient-based Post-Training Quantization (GPTQ), each reproduce none of the planted records, while the calibration-corpus-free GGUF Q4_K_M format reproduces 5.3% of them. Tracked across five open models with 0.5-7 billion parameters, AWQ leaks least at every size and in both families, with little accuracy loss at 3-7 billion. Controlled experiments associate the difference with calibration-induced rounding error in channels involved in rare-token prediction. Choosing the 4-bit method is therefore a deployment-time privacy decision, not only a question of speed and quality.
[AI-160] Do Existing Preconditioners Improve Biomedical Tabular Foundation Learning? An Empirical Study on TabPFN Optimization
链接: https://arxiv.org/abs/2609.25013
作者: M. Sajid,Pinki Khatun,M. Tanveer
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Tabular foundation models have recently shown strong potential for structured biomedical data analysis. Among them, TabPFN has emerged as an effective approach for low-data tabular classification tasks. However, the impact of optimization and preconditioning strategies on biomedical fine-tuning remains largely unexplored. In this work, we present a comprehensive empirical investigation of five AdamW-based preconditioning strategies for fine-tuning TabPFN v2.5 on 59 biomedical datasets spanning Alzheimer’s disease, breast cancer, schizophrenia, significant memory concern (SMC), KEEL biomedical datasets, and UCI biomedical benchmarks. The evaluation considers predictive performance, computational efficiency, and statistical significance analysis. Experimental results demonstrate that the original AdamW optimizer consistently achieves the best overall performance and statistical ranking, while existing curvature-aware preconditioners fail to provide reliable improvements across diverse biomedical learning scenarios. The findings suggest that generic preconditioning approaches may not adequately capture the optimization characteristics of biomedical tabular learning, motivating the development of biomedical-aware preconditioners specifically tailored for healthcare-oriented tabular foundation models.
[AI-161] Agent ic Algorithm Engineering: Improving Shared-Memory Exact Minimum Cuts
链接: https://arxiv.org/abs/2609.07204
作者: David A. Bader,Adil Chhabra,Ernestine Großmann,Monika Henzinger,Alexander Noe,Christian Schulz
类目: Artificial Intelligence (cs.AI); Distributed, Parallel, and Cluster Computing (cs.DC); Data Structures and Algorithms (cs.DS)
备注:
Abstract:The minimum cut problem for an undirected edge-weighted graph asks us to divide its set of nodes into two blocks while minimizing the weighted sum of the cut edges. Over the last years, we engineered a range of fast algorithms for this problem. Our fastest exact algorithm uses an inexact algorithm to obtain a better bound for the problem, reductions that depend on this bound, improved data structures and parallel contraction routines. It is available in the open-source package VieCut and, on real-world instances, outperformed the previously fastest solvers by a factor of up to 2.5 sequentially and up to 12.9 when run in parallel. We improve this algorithm using agentic algorithm engineering (AAE), a methodology that we introduce here, in which autonomous large language model agents run the algorithm engineering cycle on an existing code base: they form hypotheses about where running time is lost, implement them, benchmark the result on a fixed instance set and keep or discard the change. Even though we had already tuned our algorithm by hand extensively, the agent finds significant optimizations, in particular on the DIMACS core instances: factors of 1.28 (sequential) and 1.63 (32 threads) on real-world k-cores, and 6.26 and 127 on the DIMACS core instances.
[AI-162] CIAO - Code In Architecture Out - Automated Software Architecture Documentation with Large Language Models
链接: https://arxiv.org/abs/2604.08293
作者: Marco De Luca,Tiziano Santilli,Domenico Amalfitano,Anna Rita Fasolino,Patrizio Pelliccione
类目: oftware Engineering (cs.SE); Artificial Intelligence (cs.AI); Emerging Technologies (cs.ET)
备注: Manuscript accepted for the 23rd International Conference on Software Architecture (ICSA 2026)
Abstract:Software architecture documentation is essential for system comprehension, yet it is often unavailable or incomplete. While recent LLM-based techniques can generate documentation from code, they typically address local artifacts rather than producing coherent, system-level architectural descriptions. This paper presents a structured process for automatically generating system-level architectural documentation directly from GitHub repositories using Large Language Models. The process, called CIAO (Code In Architecture Out), defines an LLM-based workflow that takes a repository as input and produces system-level architectural documentation following a template derived from ISO/IEC/IEEE 42010, SEI Views \ Beyond, and the C4 model. The resulting documentation can be directly added to the target repository. We evaluated the process through a study with 22 developers, each reviewing the documentation generated for a repository they had contributed to. The evaluation shows that developers generally perceive the produced documentation as valuable, comprehensible, and broadly accurate with respect to the source code, while also highlighting limitations in diagram quality, high-level context modeling, and deployment views. We also assessed the operational cost of the process, finding that generating a complete architectural document requires only a few minutes and is inexpensive to run. Overall, the results indicate that a structured, standards-oriented approach can effectively guide LLMs in producing system-level architectural documentation that is both usable and cost-effective.
[AI-163] opology-Stratified Materials Discovery with A Flow-Based Generative Model
链接: https://arxiv.org/abs/2609.26547
作者: Jingyi Zhou,Oyshee Chowdhury,Noah Oyeniran,Chongze Hu
类目: Materials Science (cond-mat.mtrl-sci); Artificial Intelligence (cs.AI)
备注:
Abstract:Accurate generation of crystal structures is the foundation to the discovery of high-performance materials for extreme-environment applications, such as aerospace, additive manufacturing, and fusion energy systems. Although generative modeling has emerged as a promising approach for crystal design, its performance remains limited by the complex crystal structures and diverse chemical compositions. In this work, we develop UFO-MGen, a universal flow-based generative model that learns topological features of Wyckoff representations and leverages this information to accurately generate crystals across vast structural and chemical spaces. Compared with state-of-the-art generative models, UFO-MGen achieves the highest crystal generation success rate under a rigorous multi-stability evaluation framework, the highest SUN (stable, unique, novel) rate, and a remarkable extrapolation capability that has not been reported by previous models. Furthermore, a fine-tuning module is implemented to UFO-MGen for property-constrained crystal generation, enabling the inverse materials design toward target properties. The UFO-MGen opens a new avenue for accelerated materials discovery and providing a foundation for universal materials intelligence.
[AI-164] Not Quite My Tempo: Voice Activity-aware Speech Synthesis for Lip-Synchronous Dubbing INTERSPEECH2026
链接: https://arxiv.org/abs/2609.26486
作者: Alejandro Pérez-González-de-Martos,Florian Lux,Angelina Elizarova,Milana Shkhanukova,Andreas Kellner,Mattia Antonino Di Gangi
类目: Audio and Speech Processing (eess.AS); Artificial Intelligence (cs.AI)
备注: accepted at Interspeech 2026
Abstract:Automatic lip-synchronous dubbing requires a speech synthesis model to generate alternating voice and silence patterns in the target language that match the timing of the source clip precisely to ensure an optimal viewing experience. Prior works address this problem by conditioning the speech synthesis process on lip movements extracted from the video signal. In this work, we condition the speech generation on a binary voice-activity signal, which has a lightweight representation and can be produced in multiple ways. We show that the model follows the voice-activity signal with high accuracy while maintaining natural prosody and semantically appropriate pause placement within sentences, as demonstrated through extensive objective and subjective evaluations. By randomly masking this condition during training, we make the feature entirely optional during inference, allowing editors to enforce or relax lip-sync constraints when desired.
[AI-165] When Unpaired Sets Support Shared-Corruption Calibration: Moment Geometry and Two-Sample Precision
链接: https://arxiv.org/abs/2609.26209
作者: Shuheng Cao,Zhenhao Zhang,Ruiqi Chen,Renjie Cao,Siyu Zhang,Zhaoxiang Feng,Lingwei Dang,Haoyang Wu
类目: Methodology (stat.ME); Artificial Intelligence (cs.AI); Statistics Theory (math.ST)
备注:
Abstract:Collections of diverse observations often share one acquisition, processing, geometric, or channel corruption, while only an unpaired clean reference set is available. For a prescribed low-dimensional correction shared across observations, the observed and clean reference sets support inference only through the response of fixed moments. We formulate this problem as two-sample moment calibration and report a rank-aware information state combining local rank, scaled moment sensitivity, source-separated covariance, and a moment compatibility residual. Full rank gives local moment identifiability, whereas kernel directions remain unresolved to first order. A unified linearization separates observed-set and reference-set uncertainty. Under covariance weighting, the weakest scaled singular value determines worst-direction asymptotic amplification. For an orientation-preserving planar-similarity correction shared across observations, ensemble centroids and a nonzero third-order complex moment yield closed-form global population identification of translation, rotation, and isotropic scale under matched-population and no-clipping assumptions. Controlled validation tests the predicted N^-1 and \sigma_\min^-2 laws, Gaussian efficiency, and interval coverage. Bounded applications report color corrected-output quality, channel magnitude-response calibration, and a separate paired geometric de-beautification result. The framework therefore reports missing or weak information instead of treating every fitted correction as identified.
[AI-166] SE-MSB: End-to-End Unpaired Speech Enhancement using Mamba Schrödinger Bridges
链接: https://arxiv.org/abs/2609.26000
作者: Andreas Bagge,Andreas Nymand,Michael Riis Andersen,Bjørn Sand Jensen
类目: Audio and Speech Processing (eess.AS); Artificial Intelligence (cs.AI); Sound (cs.SD)
备注: 15 pages, 4 figures, 4 tables
Abstract:Speech enhancement (SE) models typically rely on supervised learning with paired data examples where clean speech is synthetically degraded. This paradigm limits performance in real-world scenarios where the target environment’s specific acoustic characteristics are unknown. We propose a fully unpaired SE framework that uses principled Diffusion Schrödinger Bridges (DSB) to learn a stochastic transport process between a clean and a degraded speech distribution. Algorithms for learning transport maps are computationally heavy since they require simulating differential equations during training, usually at each training step. Therefore, we propose using a high-efficiency Mamba Diffusion Model designed for end-to-end waveform processing. We compare against state-of-the-art methods for speech enhancement, both paired and unpaired, as well as a classical signal processing algorithm. Experimental results show that we are on par or better than the baselines while being orders of magnitude faster during inference. Furthermore, we show that the flexibility of the DSB formulation allows our model to generalize across SE tasks, offering a robust and efficient solution for real-world speech restoration.
[AI-167] Risk-Aware Online Conformal State Probing
链接: https://arxiv.org/abs/2609.25889
作者: Pietro Talli,Petar Popovski,Osvaldo Simeone
类目: ignal Processing (eess.SP); Artificial Intelligence (cs.AI)
备注:
Abstract:AI-based autonomous agents, typically hosted at data centers, must acquire state information from robots or edge devices in order to issue informed control decisions. Managing uncertainty about the state is particularly consequential in safety-critical settings, in which average-case guarantees are insufficient. In this context, we study a sequential decision maker process that jointly decides which actions to take and when to probe given access to an arbitrary state prediction model. We propose online conformal state probing (OCSP), an action and probing policy that certifies worst-case reliability levels without relying on distributional assumptions. OCSP is designed to provably control the missed query error (MQE), i.e., the fraction of instances where probing would have been beneficial, while minimizing the probing rate. OCSP can be applied to existing pre-trained value-based control policies without requiring retraining or fine-tuning. We validate OCSP through numerical simulations to verify theoretical guarantees and to assess performance trade-offs as a function of the calibration of the state predictor.
[AI-168] When Quantum Meets AI: Quantum Methods for Machine Learning and Machine Learning Methods for Quantum Systems
链接: https://arxiv.org/abs/2609.25641
作者: Tak Hur
类目: Quantum Physics (quant-ph); Artificial Intelligence (cs.AI)
备注: 171 pages, 28 figures, PhD thesis
Abstract:This thesis studies the intersection of quantum computing and artificial intelligence in two directions: quantum methods for machine learning and machine learning methods for quantum systems. For quantum machine learning, Neural Quantum Embedding learns data representations that increase the trace distance between embedded class ensembles, lowering an embedding-dependent bound on empirical risk and improving classification on noisy quantum hardware. A training objective based on the Hilbert-Schmidt inner product extends this approach to deterministic quantum computation with one qubit (DQC1) and is demonstrated on an NMR quantum processor. A margin-based generalization analysis then connects quantum neural network performance to quantum state discrimination. In the studied benchmarks, margin distributions predict generalization more reliably than parameter-count metrics. For quantum systems, a Mamba-based neural decoder for surface codes matches a reproduced Transformer baseline in memory experiments while reducing inference-cost scaling from quartic to quadratic in code distance. Under an explicit decoder-induced-noise model, it achieves lower logical error rates and a higher effective threshold. For neural quantum states, stochastic reconfiguration is interpreted as tangent-space ridge regression, with its diagonal shift controlling the bias-variance trade-off under finite Monte Carlo sampling. Multi-shift stochastic reconfiguration reduces checkpoint-local validation residuals and update variance relative to fixed-shift SR, at additional computational cost. Together, these contributions show how learned representations, statistical control, and hardware constraints shape the exchange between quantum computing and machine learning.
[AI-169] West-WRF AI 2-km: High-Resolution Prediction of Integrated Vapor Transport and Precipitation
链接: https://arxiv.org/abs/2609.25512
作者: Nazak Rouzegari,Vesta Afzali Gorooh,Agniv Sengupta,Phu Nguyen,Kuo-Lin Hsu,Amir AghaKouchak,Soroosh Sorooshian,F. Martin Ralph,Luca Delle Monache
类目: Atmospheric and Oceanic Physics (physics.ao-ph); Artificial Intelligence (cs.AI)
备注:
Abstract:We introduce a stretched-grid artificial intelligence (AI) weather forecasting model with 2-km resolution over the western United States and part of the Northeast Pacific and approximately 31-km resolution elsewhere globally. Forecasting over the western U.S. is challenging because complex topography and atmospheric rivers (ARs) strongly influence orographic precipitation. West-WRF AI 2-km builds on a global model pretrained with a 40-year European Centre for Medium-Range Weather Forecasts Reanalysis v5 (ERA5) dataset and is fine-tuned with the Center for Western Weather and Water Extremes (CW3E) 2-km regional reanalysis to produce autoregressive 6-hourly forecasts of precipitation and integrated vapor transport (IVT). Forecasts are evaluated over winters 2020-2023 using gridded precipitation observations, rain gauges, and AR Reconnaissance dropsondes and are benchmarked against coarser-resolution AI forecasts and regional and global numerical weather prediction (NWP) systems. West-WRF AI 2-km reproduces observed precipitation-intensity distributions, retains fine-scale spectral variability, and produces sharper narrow coastal precipitation bands and localized, terrain-sensitive extremes. Its broader-scale performance remains comparable to coarser-resolution configurations while preserving large-scale skill despite higher resolution. Dropsonde verification shows lower errors and improved categorical skill at the most extreme IVT threshold. Overall, West-WRF AI 2-km provides its greatest value for localized precipitation extremes and intense AR-related moisture transport.
[AI-170] ransformer-Informed Trajectory Optimization for Relative Motion in Cislunar Orbits
链接: https://arxiv.org/abs/2609.25460
作者: Walter J. Manuel,Yuji Takubo,Simone D’Amico
类目: Optimization and Control (math.OC); Artificial Intelligence (cs.AI)
备注: 2026 AAS/AIAA Astrodynamics Specialist Conference
Abstract:Autonomous spacecraft guidance and control requires a fast solution to non-convex trajectory optimization, which can be accelerated by providing a near-optimal initial guess to an optimization protocol, i.e., warm-starting. A robust warm starting method is especially useful for rendezvous, proximity operations, and docking (RPOD) in cislunar space, where the underlying dynamics become severely nonlinear and chaotic compared to those in Earth orbit, especially at perilune. This paper extends the Autonomous Rendezvous Transformer (ART), a transformer-based warm-start trajectory generation method, to cislunar RPOD scenarios for the first time. To accurately and reliably solve the nonconvex optimal control problems (OCPs) posed by these scenarios, a new and enhanced version of ART, ART-TWIN (Two-Way INference), is introduced. Inspired by forward-backward shooting methods used in other trajectory design applications, ART-TWIN autoregressively generates two arcs, one from the initial state and one from the desired terminal state, that are patched together at the midpoint of the timeseries. When evaluated on a set of simulated rendezvous scenarios that are initialized at perilune, ART-TWIN is demonstrated to substantially accelerate convergence and increase feasibility guarantees when used as a warm-start to sequential convex programming (SCP), compared to convex relaxations and the original ART. These results illustrate the necessity of ART-TWIN’s dual-arc generation to enable the viability of and gain benefits from using transformer-based warm-start methods in the most challenging areas of the cislunar dynamical regime.
[AI-171] PICPIs: Prediction-Interval-Conditional Prediction Intervals
链接: https://arxiv.org/abs/2609.25388
作者: Xuelin Yang,Baihe Huang,Yilong Hou,Guido Imbens,Michael I. Jordan
类目: Machine Learning (stat.ML); Artificial Intelligence (cs.AI); Machine Learning (cs.LG); Statistics Theory (math.ST); Methodology (stat.ME)
备注: 45 pages, 6 figures
Abstract:A classical question in statistics is which observable quantities to condition on when drawing inferences about unobservable targets. For conformal prediction in nonparametric uncertainty quantification, standard marginal validity offers limited resolution at the prediction values on which decisions are based, and fully conditional guarantees with respect to the covariates are provably unattainable. We address this gap by introducing a prediction-based conditioning framework that we refer to as Prediction-Interval-Conditional Prediction Intervals (PICPIs). Formally, a PICPI is an interval I satisfying a self-consistency condition: \mathbbE [Y \mid p(X) \in I] \in I, for predictive model p , contextual covariate X , and outcome Y . Thus, an interval simultaneously defines a stratum of prediction values and certifies that the mean outcome in that stratum lies in the same interval. This self-consistency condition yields data-adaptive strata without altering the original prediction. Such intervals can be constructed using practical algorithms. Under regularity of the prediction distribution, the constructed intervals cover all but an arbitrarily small fraction of prediction values and have widths that decrease at rate n^-1/3 , up to logarithmic factors and the prediction error. Moreover, identifying these locally calibrated intervals can, in turn, inform downstream decision-making. We derive inference procedures for PICPIs in probabilistic prediction and multi-class classification, accompanied by theoretical guarantees. Empirical results are provided that compare PICPIs with existing interval-based baselines.
[AI-172] owards Sustainable Magnetic Resonance Imaging: Insights from long-term high-resolution energy recordings across an entire scanner fleet
链接: https://arxiv.org/abs/2609.25150
作者: Florian Leonhard Raab,Fiona Mankertz,Nour Maalouf,Josephine Berger,Andreas Lingg,Reza Dehdab,Sebastian Werner,Judith Herrmann,Andreas Brendlin,Sebastian Gassenmaier,Suhas Siddaramu,Fabian Wagner,Julian Wohlers,Shreeja Varadarajan,Gurlal Singh,Jens Gühring,Rainer Schneider,Konstantin Nikolaou,Saif Afat,Thomas Küstner
类目: Medical Physics (physics.med-ph); Artificial Intelligence (cs.AI)
备注:
Abstract:Magnetic resonance imaging (MRI) is among the most energy-intensive diagnostic modalities in healthcare, yet its energy consumption and the factors influencing it remain insufficiently understood. This study aims to establish a comprehensive baseline of MRI energy consumption by characterizing energy demand across a scanner fleet, examining scanner utilization and operating patterns in clinical practice. Concurrently, it investigates the relationships between energy consumption and relevant operational and acquisition features. On average, a single MRI measurement consumed 0.43 kWh, while a complete examination consumed 13.50 kWh. In general, substantial differences in energy consumption were observed between MRI scanners and their corresponding operating modes (scan, idle, and eco-power mode). These variations may be related to differences in scanner operating patterns, employed examination protocols, and their resulting acquisition parameters. Idle and eco-power modes accounted for more cumulative energy consumption than active scanning. However, these energy shares should always be interpreted in relation to scanner occupancy, as utilization patterns strongly influence the distribution of energy across those operating modes. Lastly, linear regression analysis showed that energy consumption was more strongly associated with scan duration than with average power, suggesting that scan duration may be an important factor influencing MRI energy consumption. … Subjects: Medical Physics (physics.med-ph); Artificial Intelligence (cs.AI) Cite as: arXiv:2609.25150 [physics.med-ph] (or arXiv:2609.25150v1 [physics.med-ph] for this version) https://doi.org/10.48550/arXiv.2609.25150 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
[AI-173] Rachel: A general-purpose language model directs and revises retrosynthetic routes
链接: https://arxiv.org/abs/2609.25118
作者: Qisheng Li,Shunchao Jiang,Chen Qi,Xin Su,Da Han,Guangyong Chen
类目: Chemical Physics (physics.chem-ph); Artificial Intelligence (cs.AI)
备注: 61 pages
Abstract:Retrosynthetic planning advances through decisions that reshape the remaining chemical problem: a locally plausible disconnection can leave precursors whose chemoselectivity constraints complicate the rest of the route. Existing planners often channel model proposals through search or template procedures, leaving open whether a general-purpose large language model (LLM) can itself sustain and revise route strategy. We developed Rachel, a stateful environment that executes and checks LLM-directed chemistry but prescribes neither a search policy nor a stopping rule. Without supplied reference routes or route-level solutions, GPT-5.5 achieved strict closure for 111 of 120 PaRoutes120 targets and 24 of 25 targets in the separate RF25 difficult-target cohort. RF25 was drawn largely from studies published after GPT-5.5’s reported knowledge cutoff. Closure required complete routes and independent source resolution of every terminal precursor after planning. On a shared PaRoutes subset, forward-model support exceeded that of most comparator methods, and Rachel received the highest mean overall route score from both method-blinded LLM evaluators. Recorded trajectories showed continued model-proposed chemistry, with revised strategies carried into subsequent steps. Replacing LLM route decisions with fixed policies reduced strict closure to 6-15/120 despite continued local chemical execution; restricting planning support also reduced closure in RF25. Within Rachel, a general-purpose LLM coordinated successive chemical choices and revised its strategy as earlier decisions reshaped the remaining problems.
[AI-174] Physics-guided deep metric learning with continuous time embeddings for open-world radar pulse de-interleaving
链接: https://arxiv.org/abs/2609.25057
作者: Vikas Agnihotri,Jasleen Kaur
类目: ignal Processing (eess.SP); Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
备注: 13 pages
Abstract:Radar pulse de-interleaving is a foundational Electronic Support Measures (ESM) task that aims to separate chronologically interleaved pulse streams from multiple non-cooperative transmitters under unknown emitter cardinality in dense, contested electromagnetic environments. Classical histogram transforms and closed-world deep classifiers degrade under severe pulse loss, agile Pulse Repeti tion Interval (PRI) modulation, and spurious clutter. In this paper, we systematically characterise continuous temporal representations and physics-guided model selection in deep metric learning for open-world radar de-interleaving. Building on the transformer-based metric-learning framework for open-world deinterleaving introduced by Gunn et al. [1], we introduce a continuous Time-of-Arrival (ToA) sinusoidal positional encoding that directly models physical inter-pulse durations rather than ordinal token indices, a design choice that contrasts with [1], who found ordinal positional encodings provided no benefit and omitted them entirely. Neural network parameters are optimised solely via Supervised Contrastive (SupCon) learning, while scale-aware physical domain priors based on PRI Consistency and Angle-of-Arrival (AoA) continuity serve as physics-guided validation and checkpoint-selection criteria operating on unsupervised HDBSCAN cluster assignments.
[AI-175] Financial sentiment analysis using FinBERT with application in predicting stock movement
链接: https://arxiv.org/abs/2306.02136
作者: Tingsong Jiang,Qingyun Zeng
类目: atistical Finance (q-fin.ST); Artificial Intelligence (cs.AI); Computational Engineering, Finance, and Science (cs.CE)
备注: Large revisions with metric typos fixed. Comments welcome!
Abstract:In this study, we integrate sentiment analysis within a financial framework by leveraging FinBERT, a fine-tuned BERT model specialized for financial text, to construct an advanced deep learning model based on Long Short-Term Memory (LSTM) networks. Our objective is to forecast financial market trends with greater accuracy. To evaluate our model’s predictive capabilities, we apply it to a comprehensive dataset of stock market news and perform a comparative analysis against standard BERT, standalone LSTM, and the traditional ARIMA models. Our findings indicate that incorporating sentiment analysis significantly enhances the model’s ability to anticipate market fluctuations. Furthermore, we propose a suite of optimization techniques aimed at refining the model’s performance, paving the way for more robust and reliable market prediction tools in the field of AI-driven finance.
机器学习
[LG-0] EquivSVA: A Formally Verified Dataset of Behavioral Assertions Across Equivalent RTL Implementations
链接: https://arxiv.org/abs/2609.26751
作者: FNU Aditi
类目: Machine Learning (cs.LG)
*备注: EquivSVA is a 9-page paper by FNU Aditi introducing a formally verified dataset of 120 behavior families, 480 RTL implementations, 914 gold properties, and 360 mutants across 12 categories, featuring 2 figures, 5 tables, and a case study evaluating Qwen2.5-Coder-7B-Instruct
Abstract:Large language models are increasingly used to generate SystemVerilog Assertions from natural-language specifica- tions and register-transfer-level designs. Existing datasets and benchmarks support important goals such as large- scale training, formal evaluation, specification-to-assertion generation, and mutation-based testing. A complemen- tary need is to study whether a generated assertion cap- tures externally observable behavior or depends on inci- dental details of one RTL implementation. We present EquivSVA, a formally verified dataset organized around behavior families. Each family contains four structurally distinct RTL implementations of the same externally ob- servable behavior, shared interface-level gold properties, three controlled mutants, and formal-validation evidence. EquivSVA contains 120 behavior families across 12 cat- egories, 480 reference RTL implementations, 914 gold properties, and 360 mutants. Every final family passes a fixed 17-job validation suite covering RTL equivalence, gold-property proofs, property reachability, mutant dis- tinguishability, and gold-property checks on mutants. We also provide fixed family-safe train, development, and test splits. As a small demonstration of the analyses en- abled by the dataset, we evaluate the publicly released, Apache-2.0-licensed Qwen2.5-Coder-7B-Instruct model on the held-out test split. Of 293 interface-only generated properties, 93 are formally sound, and the number of sound properties varies across equivalent implementations for 14 of 24 test families. These results illustrate how behavior-family organization can support controlled stud- ies of assertion-generation robustness without requiring changes in intended functionality. The dataset, generators, validation scripts, and case-study artifacts are publicly released at this https URL.
[LG-1] Diffusion-Induced Spatial Attention Overlapping Community Detection
链接: https://arxiv.org/abs/2609.26737
作者: Kosti Koistinen,Vesa Kuikka,Joni Herttuainen,Matthew Hendren,Brian Holt,Kimmo K. Kaski
类目: ocial and Information Networks (cs.SI); Machine Learning (cs.LG)
*备注:
Abstract:Detection of overlapping communities is essential for modelling networks in which nodes participate simultaneously in multiple structural or functional groups. Existing graph neural network approaches commonly rely on local message passing, which can obscure community boundaries through smoothing and limit the representation of structurally relevant long-range dependencies. We introduce Diffusion-Induced Spatial Attention Community Detection (DISCO), a deep-learning framework that combines a structural prior derived from influence spreading dynamics, sparse multi-head attention, and non-negative community-affiliation learning. The prior identifies candidate interactions beyond immediate graph neighbours and biases attention according to their structural proximity, while a Bernoulli-Poisson edge-reconstruction objective enables overlapping community inference from node attributes and structural profiles, or both. Benchmark experiments show that DISCO performs competitively against established graph convolutional and graph attention approaches across different input configurations. To demonstrate its practical applicability, we present a proof-of-concept cybersecurity use case in which changes between community assignments inferred from consecutive communication-network snapshots provide an interpretable anomaly signal. Temporal community similarity identifies structural deviations, while node-level contributions help locate the devices associated with them. DISCO therefore provides both a flexible method for overlapping community detection and a foundation for analysing structural change in dynamic networks.
[LG-2] MAGIC: Mixed-Granularity Agent Graphs via Incremental Construction with Dense-Reward Reinforcement Learning
链接: https://arxiv.org/abs/2609.26667
作者: Kairui Yang,Ziheng Yi,Xunkai Li,Minghao An,Zhanke Liu,Zekai Chen,Rong-Hua Li
类目: Machine Learning (cs.LG)
*备注:
Abstract:Collaboration topology shapes both the performance and execution cost of LLM-based multi-agent systems. Because tasks differ in complexity and required capabilities, recent approaches generate task-specific collaboration graphs that specify agent participation and information flow. However, representative topology generators use either individual agents or predefined groups throughout an organization, overlooking differing collaboration needs across subtasks. Our key insight is to select granularity locally for each functional role, combining fine-grained control with reusable collaboration patterns within one organization. Learning such organizations requires exploring a combinatorial construction space with limited intermediate feedback from final-answer rewards. Therefore, we propose MAGIC, a dense-reward reinforcement learning framework for mixed-granularity graph generation. Specifically, MAGIC constructs a mixed-granularity agent graph by sequentially selecting a functional role, instantiating it as a single agent or reusable group, and connecting it to existing units. We directly optimize the construction policy using returns from trajectories sampled under the current policy and use potential-based reward shaping to provide intermediate feedback from probe-based utility and structural signals while preserving the cumulative task reward. MAGIC outperforms state-of-the-art baselines across eight benchmarks and demonstrates strong inference efficiency in our efficiency study.
[LG-3] Gap-Free Streaming PCA Beyond Rank-One Updates: Near-Optimal Rates and Applications to Differential Privacy
链接: https://arxiv.org/abs/2609.26508
作者: Anming Gu,Syamantak Kumar,Kevin Tian,Chutong Yang
类目: Machine Learning (cs.LG); Data Structures and Algorithms (cs.DS); Machine Learning (stat.ML)
*备注:
Abstract:Streaming principal component analysis (PCA) seeks to recover a leading spectral subspace in a single pass over a data stream. We give a new analysis of the ubiquitous Oja’s algorithm [Oja82] for the most general, gap-free variant of this problem, where no eigengap assumptions are made on the underlying mean matrix, complemented by a nearly-matching lower bound. Prior works achieving near-optimal rates for streaming PCA either required gap assumptions [JJK+16, HNWW21], or were limited to rank-one updates [AZL17, Lia23]. Our proof only uses a second moment bound on the individual stochastic updates, bypassing the almost sure bounds needed by prior near-optimal analyses, and the analogous offline matrix Bernstein bound. We also extend our result to a Rayleigh quotient notion of approximate PCA, addressing an open question of [JJK+16]. As our main application, we give gap-free differentially private PCA guarantees for sub-Gaussian data, settling Conjecture 1.1 of [Bro26] up to logarithmic factors.
[LG-4] Can We Predict Anomaly Detection Performance from Embedding-Space Geometry?
链接: https://arxiv.org/abs/2609.26460
作者: Kevin Wilkinghoff,Zheng-Hua Tan
类目: Machine Learning (cs.LG)
*备注:
Abstract:Anomaly detection systems are often trained using normal data alone, while model selection and evaluation typically require labeled anomalies. We study whether anomaly detection performance can be predicted without access to anomalous data. For kNN-based detectors, we derive a lower bound on the area under the ROC curve (AUC) that relates detection performance to the separation between inlier and outlier scores and to their respective variances. Under a local scaling model, we use this bound to characterize how density variation, intrinsic-dimensional heterogeneity, and cross-domain mismatch contribute to score variability. We then investigate anomaly-free model selection and show that inlier score variance alone does not reliably predict performance across different representations. To address this limitation, we introduce simple pseudo-anomaly probes that provide a reference for estimating relative score separation. Experiments on the DCASE 2022-2025 benchmarks, spanning four embedding models and 208 candidate systems, show that pseudo-anomaly-based estimators substantially improve anomaly-free model selection. In particular, diverse pseudo-anomalies enable anomaly-free model selection to outperform conventional development-set selection under domain shift. These results show that embedding-space geometry contains predictive information about anomaly detection performance while also highlighting the representation-dependent nature of inlier-only performance estimates.
[LG-5] One-Step Generative Surrogate Models via Block-Triangular Joint Drifting
链接: https://arxiv.org/abs/2609.26435
作者: Nicholas Geissler,Shreya Jha,Ricardo Baptista,Benjamin Peherstorfer
类目: Machine Learning (cs.LG); Numerical Analysis (math.NA)
*备注:
Abstract:Drifting provides a direct route to one-step generative models, but applying it directly to stochastic transition modeling requires multiple samples of the next state conditioned on the same current state. Standard trajectory data, however, typically provide only one realized next state for each observed current state and therefore do not provide an empirical approximation of the corresponding conditional distribution over possible next states. We introduce block-triangular joint drifting, which instead applies a projected drift field to the empirically accessible joint distribution of consecutive states. Importantly, the block-triangular architecture preserves the current-state marginal while making its second component a direct sampler of the conditional distribution of possible next states. The resulting surrogate generates stochastic trajectories with one model evaluation per time step, without auxiliary generative steps between time steps. Numerical experiments demonstrate accurate marginal and trajectory-dependent statistics and favorable accuracy-cost tradeoffs compared with deterministic, diffusion-, flow-, and distillation-based generative surrogate models.
[LG-6] OMatG-flash: An All-Atom Flow Map with Reinforce Adjoint Matching for Scalable Materials Discovery
链接: https://arxiv.org/abs/2609.26402
作者: Thomas Egg,Harry Winston Sullivan,Ellad B. Tadmor,Stefano Martiniani
类目: Machine Learning (cs.LG)
*备注: 27 pages, 5 figures
Abstract:The discovery of novel inorganic materials drives technological breakthroughs in critical fields such as computing and energy storage. Generative AI has promised to accelerate the materials discovery pipeline, but state-of-the-art flow and diffusion models remain bottlenecked by the cost of proposing candidate materials. To address this, we introduce OMatG-flash, an all-atom flow map for inorganic crystal structure prediction (CSP) and de novo generation (DNG). OMatG-flash is a Pareto-optimal inference engine for materials, sampling candidate materials with an order of magnitude fewer inference steps and less wall-clock time than existing flow and diffusion models while demonstrating benchmark performance on par with the state-of-the-art. To enable post-training fine-tuning we apply Reinforce Adjoint Matching to flow maps, further improving match rates and RMSE on the unconditional CSP task. OMatG-flash showcases the potential of flow maps to accelerate generation of high-quality candidate inorganic materials and demonstrates a step forward in sample throughput necessary for data-hungry materials discovery workflows.
[LG-7] Double Descent and Malign Overfitting in Diffusion Models
链接: https://arxiv.org/abs/2609.26392
作者: Raphaël Urfin,Tony Bonnaire,Giulio Biroli,Marc Mézard
类目: Machine Learning (cs.LG); Disordered Systems and Neural Networks (cond-mat.dis-nn)
*备注: 44 pages, 17 figures
Abstract:Conventional wisdom in deep learning holds that overparameterization—having more parameters p than training samples n —is benign: larger models generalize better and, even without regularization, interpolating models generalize well, the test error following a double-descent curve. One might expect the same benign overfitting for diffusion models, whose training reduces to regression, i.e. to minimizing a quadratic score-matching loss. Yet the opposite is observed: overfitting here is catastrophic, driving the model into a memorization regime. We resolve this paradox by combining experiments on U-Nets trained on CelebA with a random-features model for which we derive closed-form learning curves. We show that with a fixed number m of noise realizations per training sample, an interpolation peak does occur, but at p\sim nm rather than at p\sim n as in standard regression. The rise of the test loss, however, sets in much earlier, at p\sim n , independently of m . This overfitting is malign because, although the implicit regularization of training is fully at work, it drives the model toward the empirical score, which memorizes the training set, rather than toward the true score. A bias-variance decomposition pinpoints the mechanism: the bias of the score estimator starts to grow at p\sim n ; past the peak the variance decays, as in regression, whereas the bias keeps growing and both saturate at a large value. Since diffusion models are trained with m\gg1 , the peak is pushed to very large model sizes, and therefore sit on the rising branch that precedes it, where malign overfitting is already in play. Nevertheless, overparameterization remains beneficial when paired with regularization: in the random-features theory and in U-Net experiments, optimally regularized large models—via a ridge penalty or early stopping, respectively—outperform any unregularized models.
[LG-8] Learning to Defer with Guidance on Real World Medical Data MICCAI2026
链接: https://arxiv.org/abs/2609.26384
作者: Emma Sun,Joshua Strong,Alison Noble
类目: Machine Learning (cs.LG)
*备注: Accepted at HAIC workshop, MICCAI 2026
Abstract:Medical image interpretation is high-volume and time-consuming, and while AI interpretation can reduce workload, fully autonomous deployment carries potential safety concerns and low specificity may in practice lead to increased clinician workload. Learning to Defer (L2D) addresses this by selectively routing cases between autonomous prediction and human experts by learning from input features and AI model and human performance. While theoretical guarantees have been proven for L2D, its performance has not been validated on real-world medical datasets with human reader annotations. We evaluate the predictor-rejector formulation of two-stage L2D, where the AI predictor model is fixed and separate from the trainable routing or rejector model, on Collab-CXR, a multilabel chest X-ray dataset with multiple human annotations per case. This is the first work to look at L2D in the context of real-world medical imaging data with human annotations. We further introduce a new setup, L2D with Guidance, where the decision space is extended to three choices: predict autonomously, defer to a human expert, or defer to a human expert and provide AI guidance. We compare multiple rejector architectures and loss functions, and different input feature availabilities. This is reproduced on two larger datasets, VinDr-CXR and CheXpert. Our results show that two-stage L2D with Guidance outperforms classic two-stage learning to defer, as well as human-alone, AI-alone and AI-guided human baselines. Notably, this performance is achieved with simpler loss functions compared to formally defined L2D surrogate loss functions in current literature.
[LG-9] HYDRA: Proactive Android Malware Drift Adaptation via Hierarchical Graph Contrastive Learning CCS2026
链接: https://arxiv.org/abs/2609.26352
作者: Han Chen,Hanchen Wang,Hongmei Chen,Lu Qin,Wenjie Zhang,Ying Zhang
类目: Cryptography and Security (cs.CR); Machine Learning (cs.LG)
*备注: Accepted at ACM CCS 2026. Author’s version with full appendix. 17 pages
Abstract:Concept drift, driven by the rapid evolution of Android malware, severely degrades the performance of machine learning detectors. Current adaptation strategies are often reactive, responding only after performance has dropped and imposing a significant manual annotation burden, or they are proactive but rely on unstable adversarial training and incomplete, single-level graph representations. To overcome these limitations, we propose HYDRA (Hybrid Drift Adaptation), a proactive adaptation framework that learns drift-invariant representations from hierarchically structured data. HYDRA first models applications using a hybrid graph structure, combining fine-grained Control Flow Graphs (CFGs) and coarse-grained Function Call Graphs (FCGs) to capture comprehensive behavioral patterns. It then introduces a novel cross-domain contrastive learning objective that aligns historical (source) and new (target) data distributions. By generating pseudo-labels for unlabeled target samples, our method pulls representations of semantically similar applications together, regardless of their domain, within a single, stable optimization process. This approach unifies feature learning and domain alignment, eliminating the need for complex adversarial objectives. Extensive experiments on large-scale, time-ordered malware datasets demonstrate that HYDRA achieves substantially lower False Negative and False Positive Rates than state-of-the-art baselines while requiring up to 87.5% fewer labeled samples. Our work thus offers a robust and efficient solution to combat concept drift in security applications.
[LG-10] Disaggregated Quantization: Specializing LLM Prefill and Decode
链接: https://arxiv.org/abs/2609.26333
作者: Andrei Panferov,Maximilian Kleinegger,Sweta Priyadarshi,Tijmen Blankevoort,Dan Alistarh
类目: Machine Learning (cs.LG)
*备注:
Abstract:Prefill and decode reward different approaches to quantization: low-precision arithmetic accelerates prompt processing, while compact weights reduce memory traffic during generation. We propose “disaggregated quantization” (DQ), which specializes computation formats, weights and storage placement to both of these phases. On Qwen 3 and Gemma 3, removing activation quantization specifically on decode improves accuracy on decode-heavy tasks without increasing inference cost. Training separate compute-native prefill weights accelerates prompt processing relative to weight-only inference while matching or exceeding its accuracy at 2-3-bit decode on both decode-heavy and prefill-heavy tasks. With released Qwen3.8-27B GGUF decoders, training an NVFP4 prefiller improves 1-bit accuracy by 32.5 points on MMLU-Pro and 35.3 on MMMU-Pro without modifying the decode checkpoint. To accommodate the additional checkpoint on a single device, offloaded disaggregated prefill (ODP) streams its weights from SSD, amortizing loading over prompt length. On the same 27B model, ODP delivers a 1.78x time-to-first-token speedup over the weight-only baseline at 8K prompt length in this http URL. We evaluate accuracy under disaggregated serving in vLLM and further validate shared-weight format disaggregation through post-training quantization on models up to 2.8T parameters.
[LG-11] PreGS: A Parameter-Transfer-Based Multi-Expert Graph Neural Network for Node Classification
链接: https://arxiv.org/abs/2609.26310
作者: Zhicong Cai,Yinglong Zhang,Xiaoying Hong,Xuewen Xia,Xing Xu
类目: Machine Learning (cs.LG)
*备注: 12 pages, 3 figures, 7 tables
Abstract:Graph neural networks have achieved strong performance in node classification by aggregating information from graph neighborhoods. However, a single aggregation mechanism may be insufficient to capture diverse structural patterns across graph datasets. Moreover, independently training multiple structural branches can introduce substantial overhead without necessarily producing stable node representations. To address these issues, this paper proposes PreGS, a parameter-transfer-based multi-expert graph neural network framework. PreGS first pretrains a multi-head graph attention network (GAT) and transfers the linear transformation weights of its first-layer attention heads to multiple GraphSAGE experts. The transferred experts are frozen and used as complementary structural branches. The fused raw node features, GAT head representations, and GraphSAGE expert representations are fed into a multilayer perceptron (MLP), whose output is further fused with the pretrained GAT logits. Based on PreGS, we further develop PreGSv2, which introduces source-level weighting and a structural gating mechanism for adaptive multi-source feature integration. Experiments on eight public graph datasets show that PreGS and PreGSv2 achieve competitive performance against representative graph neural network baselines. Ablation, parameter-transfer, sensitivity, aggregator, visualization, and training-time analyses further validate the effectiveness and stability of the proposed framework. The code and datasets are available at this https URL.
[LG-12] Quantifying Protocol-Induced Uncertainty in Comparative Predictive-Model Evaluation: Evidence from Large-Scale Daily PM10 Forecasting
链接: https://arxiv.org/abs/2609.26288
作者: Rafael da Silva,Kiersten Monahan
类目: Machine Learning (cs.LG)
*备注: 29 pages, 8 figures, 7 tables
Abstract:Comparative studies of predictive models often end by ranking candidate models, yet these rankings depend on evaluation protocols whose influence is rarely treated as a source of uncertainty. We formalize this problem as protocol-induced ranking uncertainty and introduce a framework that compares ranking displacement caused by switching protocols with displacement produced by conventional choices within a fixed protocol. We quantify these effects using the Protocol Sensitivity Score (PSS) and a full-refit intraprotocol reference. We validate the framework in a large-scale sequential prediction study of daily PM10. Static-split and rolling-origin evaluation are compared across 425 European background stations and 365 US EPA monitors. Switching protocols produces mean PSS values of 0.801 and 0.772 and changes the selected model at 35.3% and 31.5% of stations, respectively. In Europe, intraprotocol perturbations with identical scored targets produce PSS values of 0.072 and 0.230, with winner-swap rates of 0.8% and 4.9%. Between-protocol displacement is therefore substantially larger than the selected within-protocol references. Expanding the candidate set from three to nine models increases the between-protocol winner-swap rate to 60.2% in Europe. The pattern also persists under a frozen protocol applied to held-out background and non-background stations. These results show that model-selection conclusions can depend materially on legitimate evaluation choices. We recommend reporting ranking stability under a small set of defensible intraprotocol perturbations alongside claims of model superiority. Comments: 29 pages, 8 figures, 7 tables Subjects: Machine Learning (cs.LG) Cite as: arXiv:2609.26288 [cs.LG] (or arXiv:2609.26288v1 [cs.LG] for this version) https://doi.org/10.48550/arXiv.2609.26288 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
[LG-13] On the Effect of Bit-Level Parameter Perturbations in Machine Learning and Deep Learning Models
链接: https://arxiv.org/abs/2609.26280
作者: Akanksha Raghapur,Mark Stamp
类目: Machine Learning (cs.LG)
*备注: To appear as a chapter in the book “Artificial Intelligence for Cyber Defense in Emerging Threats”, to be published by Springer by early 2027
Abstract:In this chapter, we investigate how classical machine learning models respond to small, targeted modifications in their parameters. We compare and contrast these results to analogous experiments on deep learning models. For classical learning models, we consider Hidden Markov Models (HMM) and Support Vector Machines (SVM), and for comparison, we conduct analogous experiments involving Multilayer Perceptrons (MLP) and Long Short-Term Memory (LSTM) networks. When applied to the Drebin Android malware dataset, our results show that classical models are brittle, in the sense that a limited set of selected parameters can have a dramatic effect on model behavior. In a related set of experiments, we investigate the steganographic capacity of these same learning models, that is, the proportion of bits in model parameters that can be overwritten without having a significant adverse affect on a model. We find that classical models offer limited steganographic capacity due to their compact, parameter-efficient, and relatively sensitive parameter structure. In contrast, neural networks are parameter-redundant, enabling higher steganographic capacity, where modifications can be distributed across many parameters with minimal impact on performance. These results highlight differences in how classical and neural models respond to parameter changes, with clear implications for both robustness and hidden information embedding. Overall, this work provides a framework for understanding parameter sensitivity and steganographic capacity across different classes of learning models.
[LG-14] JAMPR/L2D: scalable neural heuristic for constrained vehicle routing problems in dynamic environment
链接: https://arxiv.org/abs/2609.26275
作者: Andrew Soroka,Alex Meshcheryakov
类目: Machine Learning (cs.LG)
*备注: Automation and Remote Control accepted
Abstract:The vehicle routing problems with real-world constraints (we consider vehicles capacity limits, time windows constrains, pickup-and-delivery multi-depo — CPDPTW) pose significant computational challenges. While classical exact and heuristic methods remain effective to solve problems of small/medium size ( N\lesssim100 ), they often lack adaptability and scalability for larger logistics tasks. In this work, we show how JAMPR+/L2D RL deep learning model, proposed in to solve large CPDPTW problems can be adopted in the case of substantial changes of graph distance matrix. We test performance of JAMPR+/L2D model for medium-sized CVRP and VRPTW problems on CVRPLIB benchmarks: JAMPR+/L2D outperforms the state-of-the-art heuristic HGS in over 85% of instances, achieving improvement in objective gap. We show that the JAMPR+/L2D model trained on CPDPTW problem, generalizes well for tasks with simpler constraints (CVRP, VRPTW), for different problem sizes and for moderate changes in distance matrixes. For more substantial changes in distance matrixes, we propose here to make fast finetuning of JAMPR+: on ORTEC data (for CPDPTW) the proposed strategy remarkably reduces the objective gap without full model retraining, what will give both accuracy and rapid inference of the model in the practical routing scenarios with distance matrix changes.
[LG-15] Mode Collapse Is Cheap to Detect: A Ground-Truth-Free Pre-Flight Check for Neural Samplers
链接: https://arxiv.org/abs/2609.26272
作者: Jian Xu
类目: Machine Learning (cs.LG)
*备注:
Abstract:Neural samplers are trained against an unnormalised target \tilde\pi=e^-E with no samples from \pi , which leaves the practitioner with no way to tell whether an expensive training run has silently dropped part of the target. The diagnostics in common use are computed from the model’s own draws and are therefore confined to the model’s support: we exhibit a sampler whose self-normalised effective sample size is 0.99 while it misses 87% of the target mass. We argue that \emphdetecting missing mass is a strictly easier problem than sampling it: detection needs one point per missed basin plus a local curvature estimate, whereas correction needs the sampler retrained. We turn this into a pre-flight check that consumes a few percent of the sampler’s own training budget and uses only E , \nabla E and \nabla^2 E . On Gaussian-mixture, Many-Well and rotated anisotropic Many-Well targets with exactly computable ground truth, the check estimates the missing mass to within 10^-3 at 2.7% of training cost, where a tuned annealed SMC reference needs 70 – 280% of training cost to do worse. It also applies unchanged to a controlled-SDE sampler that has no tractable density, where ESS and the ELBO cannot be formed at all. The estimator carries a \emphself-diagnostic that, without ground truth, is conservative in the safe direction: across 60 configurations it clears 16 , of which 15 are accurate to 10^-2 or better. We are explicit about what this does and does not license: the check cheaply produces evidence of missing mass, and sometimes evidence that the search has stabilised, but it cannot certify a run, and its thresholds are heuristic. We then map the boundary of the method on a real physical landscape, LJ-13, and report where it fails and why.
[LG-16] Information-Theoretic Decoupled Prompt Tuning for Continual Learning
链接: https://arxiv.org/abs/2609.26257
作者: Yunfei Zhang,Wen Wen,Tieliang Gong,Weizhan Zhang
类目: Machine Learning (cs.LG)
*备注:
Abstract:Continual learning (CL) aims to incrementally acquire knowledge from sequential data while avoiding catastrophic forgetting. Recently, prompt tuning has attracted increasing attention as an efficient approach for adapting pre-trained models to CL tasks. However, existing prompt design paradigms commonly suffer from retrieval dependence and classifier bias, which make model adaptation sensitive to prompt selection and bias predictions toward newly arrived classes. To address these challenges, we propose Decoupled Prompt Tuning for Continual Learning (DPT4CL), which decouples the CLIP textual prompt into a task-shared prompt distribution and class-specific prompts. The task-shared prompt distribution is derived by optimizing an Information Bottleneck objective to facilitate cross-task knowledge transfer and alleviate classifier bias, while class-specific prompts enhance inter-class separability without relying on explicit prompt retrieval. Furthermore, we establish a unified excess risk bound from an information-theoretic perspective, providing theoretical support for the robust generalization and forgetting mitigation of the proposed framework. Extensive experiments on standard CL benchmarks demonstrate that DPT4CL achieves state-of-the-art performance. The source code is available at this https URL
[LG-17] Can You Delete a Year of Market Data? Machine Unlearning Against Exact Retraining Oracles
链接: https://arxiv.org/abs/2609.26242
作者: Junyi Ye
类目: Machine Learning (cs.LG); Statistical Finance (q-fin.ST)
*备注:
Abstract:When a data license expires, deleting stored records does not remove influence encoded in a trained forecaster. Machine unlearning seeks to remove this influence without retraining. We benchmark temporal unlearning with 3,200 paired references trained on all data and oracles retrained without the requested period. The grid covers five architectures, four rolling folds, five deletable years, and three experimental deletion levels on an SP 500 volatility panel. The 2020 COVID crisis year produces the largest memorization gap for every architecture. Removing it improves all three deployable models in every fold, with the largest improvement in the 2022 bear market, while the two non-deployable models respond inconsistently. The target for approximate unlearning is the oracle, not low predictive accuracy on the deleted period. In one Transformer cell, an oracle that never trained on 2020 still predicts it at an information coefficient of 0.51, compared with 0.55 for the reference; pushing predictions toward noise reduces test skill. Across twelve deployable architecture-method pairs, only TSMixer with the hinge method remains near the oracle in every fold, closing 74-118% of the reference-to-oracle gap without a measurable loss of test skill. Method rankings vary across architectures and rolling windows. Audit separation rises with prior memorization but can remain small after exact deletion. The window-level loss comparison reaches at most 0.69, and treating stock-level windows as independent inflates the absolute t-statistic by a median factor of 1.9. These results call for an explicit deletion scope, oracle validation for the relevant architecture and window, and power-aware auditing.
[LG-18] High-Order Liquid Evidence Modeling for Continuous and Subtle GNSS Spoofing Detection in Autonomous Driving
链接: https://arxiv.org/abs/2609.26231
作者: Muhammad Ayub Sabir,Junbiao Pang,Fatima Ashraf
类目: Machine Learning (cs.LG)
*备注:
Abstract:Continuous and subtle GNSS spoofing poses a serious threat to autonomous vehicles because forged positions may remain locally plausible while gradually becoming inconsistent with vehicle motion observed by non-GNSS onboard sensors. Existing AV-oriented detectors commonly rely on residual thresholds or feature-level classification and provide limited modeling of how weak GNSS–motion inconsistency develops and persists over time. This paper formulates subtle GNSS spoofing detection as a causal sequential evidence-modeling problem and proposes a high-order liquid evidence detector. The method first compares the displacement implied by consecutive GNSS positions with that inferred from independent onboard motion observations and converts their difference into uncertainty-normalized residual evidence. It then represents the current inconsistency, its local evolution, excess above the normal level, accumulated persistence, and displacement validity as causal weak evidence. These cues are mapped into instantaneous, evolutionary, and persistent latent states, aligned through a bounded Kirchhoff-inspired symmetric exchange, and combined through an explicit third-order interaction to capture their coordinated support for spoofing. To model how this coordinated evidence develops over time, second-order liquid dynamics track its memory and evolution to estimate causal spoofing probabilities, which are converted into confirmed alarms using validation-selected threshold and persistence parameters. Experiments on the AV–GPS dataset family demonstrate strong controlled and external generalization, together with clear sequential alarm behavior. On Dataset-1, the proposed detector achieves an AUROC of 0.9932 and an AUPRC of 0.9843, while obtaining the lowest false-positive rate among the learning-based baselines. Code: this https URL.
[LG-19] PatchKV: Efficient KV Cache Recovery for Dynamically Edited LLM Contexts
链接: https://arxiv.org/abs/2609.26219
作者: Guotao Yang,Rui Guo,Siwei He,Sheng Chen,Yitao Hu,Keqiu Li
类目: Distributed, Parallel, and Cluster Computing (cs.DC); Machine Learning (cs.LG)
*备注: 10 pages, 10 figures, 2 tables
Abstract:Long-running LLM agent workflows often revise interior context spans while retaining long suffixes. Although suffix tokens remain unchanged, altered causal histories and rotary positions prevent exact reuse of their offloaded key-value (KV) states. Full suffix recomputation wastes prefill work, while indiscriminate reuse propagates stale states and full-precision restoration adds data movement. We present PatchKV, a profile-guided recovery system for suffix-preserving revisions. PatchKV decomposes adjacent context versions into an exact prefix, an updated span, and an aligned suffix. It predicts an edit-local dirty region using an offline length-conditioned drift model, augments this region with sparse nonlocal blocks selected from stored attention, and block-rounds their union into a fixed repair set. The remaining suffix blocks are restored from CPU memory using frozen per-block precision tags and a fused path for dequantization, RoPE correction, and KV-page placement. Across three models and three long-context question-answering workloads, PatchKV achieves a 2.51 - 3.85\times speedup in mean resume time-to-first-token over full suffix recomputation and a 1.26 - 2.06\times speedup over CacheBlend, while matching or exceeding CacheBlend’s F1 score in six of nine settings and remaining within 1.36 points in the others.
[LG-20] MSA-CITE: A Co-Adapted LoRA Specialist Ecology for Fixed-Budget Small-Model Inference
链接: https://arxiv.org/abs/2609.26217
作者: Ruitong Li,Binjie Guo,Aisheng Mo,Guowei Su,Jie Li,Ru Zhang
类目: Machine Learning (cs.LG)
*备注:
Abstract:Compact language models are typically deployed by retaining a single post-training checkpoint and sampling it repeatedly. In this work, we challenge this practice by treating multiple discarded checkpoints as composable assets for deployment. Starting from a single Qwen3-4B backbone, we preserve four frozen LoRA branches, each derived from a different post-training trajectory. stead of drawing four generations from one branch, we allocate a fixed four-generation budget by sampling one completion from each branch. Our method, Multi-path Specialist Adaptation with Calibrated Inference-Time Evidence (MSA-CITE), processes the resulting portfolio by grouping terminal answers into equivalence classes, scoring each class via summed calibration-derived source priors, and selecting a representative under deterministic tie-breaking rules. The readout stage does not learn from evaluation results, nor does it introduce additional generations, verifiers, or reranking steps. On 200 held-out mathematics items, the four-path portfolio achieves 65.5% accuracy, compared with 62.0% for the strongest single-branch baseline. On a 100-item subject-disjoint shift, it attains 42.0% versus 40.0%. Under in-distribution conditions, the improvements over homogeneous SFT and Online-OPD repetition are robust; results against the strongest baseline and under shifted conditions are not conclusive. Our findings offer a narrow but concrete contribution: post-training branches, even without co-training, can be collectively beneficial for deployment.
[LG-21] Beyond Imitation: Auditing the Recoverability of Reasoning in Distilled Models
链接: https://arxiv.org/abs/2609.26216
作者: Ruitong Li,Binjie Guo,Aisheng Mo,Guowei Su,Han Wang,Jie Li,Ru Zhang
类目: Machine Learning (cs.LG)
*备注:
Abstract:A correct teacher solution becomes useful supervision when the receiving student can continue its reasoning. We measure this compatibility with prefix recovery: after revealing 25%, 50%, or 75% of a verified solution, we test whether the student completes it correctly. We connect recovery to the cosine conflict between cross-entropy and reverse-KL gradients over the full vocabulary. Across adjacent Qwen3 teacher-student pairs from 0.6B to 8B parameters, reverse-KL distillation delivers its most consistent mathematical and code improvements for the two students below 2B parameters. On a fixed cohort of 1,000 trajectories, average prefix recovery rises from 71.0% to 91.9% as student size increases from 0.6B to 4B, and the robust-fragile recovery gap contracts from 46.0 to 14.4 percentage points. With the teacher fixed at 8B, conflict separation falls from 0.993 to 0.233. An independent objective intervention finds the largest reverse-KL rescue on fragile trajectories. The three measurements locate the same capacity-dependent transfer regime: distribution matching has the greatest headroom when correct traces remain unevenly recoverable. Prefix recovery provides a practical diagnostic for selecting costly distribution-level distillation.
[LG-22] Bridging the Data Gap: Digital Twin as a New Paradigm for AI-based Radio Sensing
链接: https://arxiv.org/abs/2609.26214
作者: Éloi Sainte-Beuve(1, 2),Guillaume Larue(1),Louis-Adrien Dufrène(1),Quentin Lampin(1),Ali Al Khansa(1) ((1) Orange Research, (2) IMT Atlantique)
类目: Machine Learning (cs.LG); Signal Processing (eess.SP)
*备注: Accepted at EUSIPCO 2026
Abstract:We present a methodology that places a 3D digital twin (DT) of the environment as the main enabler behind the development of radio sensing at scale. The DT acts as a world model, providing geometry, materials, and transmitter/receiver placements to a ray-tracing engine that generates time-indexed channel impulse responses (CIRs) for large numbers of plausible scenes (moving people and objects, layout variants, seasonal/weather conditions, etc). From these synthetic sequences, we train a sequential neural network that maps CIR time series to spatial occupancy estimates, enabling device-free localization (DFL) without instrumented targets. We posit that sensing is best approached as an environment-conditioned learning problem: rather than seeking a single global model, we advocate training or fine-tuning local models specialized to a site-specific DT. As a first experiment, we introduce a novel State Space Model architecture, trained and evaluated across multiple room geometries. The localization performances obtained demonstrate the potential of the approach.
[LG-23] Partially Observed Sparse Graphs: The Unknown Sampling Rate is a Tail Index
链接: https://arxiv.org/abs/2609.26199
作者: Jian Xu,Delu Zeng,John Paisley,Qibin Zhao
类目: Machine Learning (cs.LG)
*备注:
Abstract:A large graph is often available only in part: a crawl stopped by its budget, a panel, a partial dump. When the sampled fraction s is known by design the total edge count follows from \hat e=e_s/s^2 and no model is needed. We treat the case where s is unknown and the population size is known. Our main result is a reduction: under a sparse exchangeable (graphex) model the expected non-isolated fraction obeys n_s/n_1\to s^1+\sigma , so the sampling rate becomes estimable once the tail index \sigma is, and substituting it back gives e_s(n_1/n_s)^2/(1+\sigma) – the same estimator, with the design quantity inferred. Estimating global edge cardinality in a sparse graph is therefore, in expectation, tail-index estimation, and the quadratic graphon estimator is the case \sigma=0 : it fails by an identity rather than by a fit ( 260% median error against 27% ). We bound the finite-size error of the substitution and show the reduction is \emphmodular in the tail-index estimator — filled with a published closed-form one it reaches 21.7% over 13 networks and 39 sampling budgets with no fitting at all. Fitting a full graphex additionally returns the degree distribution at any size and a generative object, in a representation where sparsity is a coordinate and the interpolation path is dictated rather than chosen. Two limits are exact: rank-one graphexes have transitivity fixed by the degree profile, so high-clustering graphs lie outside the class; and under snowball or random-walk crawls every method here fails, the design-based oracle worst of all ( 7.8% to 588% ).
[LG-24] Component Type Not Reconstruction Error Predicts Attention Quantization Sensitivity ICML
链接: https://arxiv.org/abs/2609.26173
作者: Kasun Dewage,Marianna Pensky,Suranadi De Silva
类目: Machine Learning (cs.LG)
*备注: Accepted as a regular paper at IEEE ICMLA 2026; to appear in the conference proceedings
Abstract:Many post-training quantization (PTQ) methods use layer-wise reconstruction, second-order proxy objectives, or activation-aware transformations to reduce quantization-induced error. Whether that error signal predicts the downstream functional impact of quantizing an individual attention projection has not been directly characterized. We sweep nine open-weight language models (1.3B–8B parameters; OPT, GPT-J, LLaMA-1/2/3, Mistral, Qwen 2.5) and quantize one attention projection at a time under round-to-nearest (RTN) and, for seven models, GPTQ at 3 and 4 bits, recording reconstruction error, perplexity change, and per-projection activation-weighted quantization error for 3,808 distinct measurements. We find: (1) within a given component type (Q, K, V, or O), reconstruction error explains less than 10% of the variance in perplexity sensitivity in 27 of 36 cases under RTN, with median R^2 = 0.044; (2) both component type and layer identity explain more variance than reconstruction error in all 9 models, with layer identity the strongest predictor in 7 of 9 models and component type strongest in the remaining 2; (3) value (V) projections are the most commonly dominant component, accounting for 38–51% of total positive Delta PPL in seven of nine models; (4) the dominant component is broadly preserved between RTN and GPTQ (5 of 7 cases); and (5) activation-weighted quantization error is a moderately better within-component predictor than reconstruction error for V projections specifically (median R^2 of 0.20 vs. 0.06). These findings indicate that relative weight reconstruction error alone is insufficient for sensitivity-aware bit allocation, and that V projections merit dedicated consideration in mixed-precision schemes.
[LG-25] Activation-Energy Pruning for Spiking Neural Networks: Unsupervised Personalization via Spike-Count Saliency
链接: https://arxiv.org/abs/2609.26167
作者: Joseph Bingham
类目: Machine Learning (cs.LG); Neural and Evolutionary Computing (cs.NE)
*备注: 38 pages, 17 figures
Abstract:Activation-energy pruning – removing weights whose product of magnitude and cumulative pre-synaptic spike count falls below a threshold – was established as an effective unsupervised personalization strategy for conventional deep neural networks~\citepBINGHAM2025101242. This paper asks what happens when the same criterion is applied to spiking neural networks (SNNs), where activation energy is not merely a useful heuristic but a literal physical quantity proportional to the metabolic cost of each synapse. The answer is surprising on three counts. First, gradient-based pruning methods that perform competitively on conventional networks (SNIP, GraSP, magnitude pruning) consistently underperform on SNNs, collapsing to near-chance accuracy by \sigma = 0.2 sparsity across all tested architectures and datasets. We trace this to a systematic incompatibility between surrogate-gradient saliency estimation and the binary spike-train representation, though we cannot rule out that alternative surrogate choices or hyperparameter settings might partially mitigate the effect. Second, activation-energy pruning applied to a neuromorphic benchmark \emphimproves over the source model at high sparsity ( 98.4 \pm 0.4% vs.\ 97.2 \pm 0.7% at \sigma = 0.8 on N-MNIST), a phenomenon with no counterpart in the conventional network setting. We interpret this result as consistent with experience-dependent cortical specialisation: removing connections active only for non-target classes may reduce cross-class interference and produce a cleaner target representation, though we note this is an interpretive analogy rather than a mechanistic demonstration. Comments: 38 pages, 17 figures Subjects: Machine Learning (cs.LG); Neural and Evolutionary Computing (cs.NE) MSC classes: 68T07 ACMclasses: I.2.6; I.5.1; J.3 Cite as: arXiv:2609.26167 [cs.LG] (or arXiv:2609.26167v1 [cs.LG] for this version) https://doi.org/10.48550/arXiv.2609.26167 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
[LG-26] Spectral Tail Interventions in Decoder-Only Language Models: Reasoning -Sensitive Weight Structure from Controlled Surgery
链接: https://arxiv.org/abs/2609.26165
作者: Ibne Farabi Shihab,Sanjida Akhter,Md Najmus Swaqeeb,Abu Sa-Adat Mohamed Moon-Im Al Ahsan,Anuj Sharma
类目: Machine Learning (cs.LG)
*备注:
Abstract:Weight-space structure often correlates with language-model behavior, but correlation alone does not establish computational involvement. We study concentrated upper spectral tails in decoder-only transformers through controlled interventions. At a fixed relative offset, we derive a finite-width conditional bound linking the inverse participation ratio of squared singular values to central pre-softmax logit kurtosis. We then define a pointwise query–key ( QK ) product-tail target and compare independent factor surgery with a product-targeted factorization that preserves native attention computation. Across three base checkpoints and five reasoning benchmarks, plus an instruction-tuned Phi checkpoint analyzed separately, the learned-tail edit is more damaging than the mean of five fixed spectrum-matched Haar controls in all 20 model–task cells. Eighteen paired contrasts remain significant after Holm correction, while two are directional but inconclusive. Product-targeted factors attain higher held-out tail-subspace fractions, providing an empirical bridge between product- and factor-level interventions. Component isolation identifies contributions from QK , value–output, and multilayer-perceptron blocks, although the theorem covers only QK . In separate studies, inverse participation precedes pooled accuracy transitions under a matched crossing rule, and residualized tail-aware low-rank adaptation (LoRA) reaches targets earlier than standard LoRA and PiSSA while final-score intervals overlap. Conclusions are restricted to the evaluated checkpoints, layers, tasks, interventions, and controls.
[LG-27] Block-Level Weight-Space Structure Persists Under Post-Training: An Empirical Study Across LLM Families ICML2026
链接: https://arxiv.org/abs/2609.26147
作者: Zhaohui Wang
类目: Machine Learning (cs.LG)
*备注: 13 pages, 11 figures. Accepted at the ICML 2026 Workshop on Weight-Space Symmetries: from Foundations to Practical Applications. OpenReview: this https URL
Abstract:Modern LLMs are deployed as families of post-trained variants (base, instruct, chat, code) derived from a shared set of pre-trained weights. We present an empirical study of how post-training transforms weight-space geometry, covering eight configurations across four architecture families (Qwen2.5, Llama-3.1/3.2, Mistral, Gemma-2). We identify a granularity gap: post-training modifies every tensor (zero of 291-339 tensors remain byte-identical, so hash-based deduplication achieves 0% savings), yet preserves block-level structure (mean cosine similarity exceeds 0.99 and relative Frobenius distance stays below 0.13). Post-training therefore acts as a structured perturbation that shifts every parameter while leaving block-level geometry intact. The property is not universal: independently trained specializations (for example, Qwen2.5-Coder) attain cosine similarity around 0.64 with the general base, indicating a disconnected region of weight space. Perturbation magnitude varies systematically with model scale, architecture family, and post-training recipe. As a practical application, we build LinkerLLM, a lazy loader that aliases shareable blocks across co-resident variants, achieving 18-48% GPU memory savings and enabling up to five 7B-parameter variants on a single 24 GB consumer GPU. Five of eight configurations retain at least 94% of the unshared variant’s quality on MMLU, ARC-Challenge, HellaSwag, and WinoGrande; the remaining three (Mistral-7B, Gemma-2-2B, Llama-3.2-1B) have one below-threshold benchmark each (87-91%), which we report transparently rather than gate the block-sharing decision on a single threshold.
[LG-28] CoEvo: Oracle-Grounded Self-Evolution of a Single Model for Multi-Step Causal Reasoning
链接: https://arxiv.org/abs/2609.26094
作者: Jian Zhang,Bingyi Wang,Yizhi Liu
类目: Machine Learning (cs.LG)
*备注: 9 pages, 2 figures
Abstract:Multi-step causal reasoning requires chaining inferences where each step constrains the next. An early error propagates silently, and a correct answer reached via flawed logic evades outcome-level detection. In specialized domains, teacher LLMs err on intermediate steps, safety constraints restrict cloud distillation, and shifting conditions demand adaptation, leaving self-evolution as the practical route. Naive self-evolution can collapse: outcome-only rewards let the model exploit distributional shortcuts, and weak self-evaluation reinforces spurious paths into stable failure patterns. We exploit a key asymmetry: generating a correct chain is hard, but verifying a single step is easy. Many high-stakes domains admit a deterministic, queryable oracle, a physics simulator or rule engine over codified constraints. It checks asserted steps without teacher-level ability and abstains beyond its rules; it can check what the model asserts, never replace it. This enables CoEvo, an oracle-grounded self-evolution framework where a single model alternates between Proposer and Solver. As Solver, the model generates competing chains; intra-group debate exposes disagreement steps, a proxy for the capability boundary, and the oracle adjudicates them into process-level supervision. As Proposer, the same model constructs progressively harder scenarios inside oracle constraints, steering the curriculum toward deep multi-hop chains. Both roles are updated jointly, so training pressure co-evolves with the model. On industrial, clinical, and legal multi-step causal reasoning benchmarks, CoEvo enables an 8B LLM to sustain self-evolution, surpassing distillation baselines and the strongest proprietary reference on path correctness (82.1% vs. 71.4%). The trained model generalizes to unseen categories and systems, preserving root-cause accuracy.
[LG-29] Fast Matrix Multiplication in fp8: Certified Coefficient Optimization and Measured Error FAST
链接: https://arxiv.org/abs/2609.26077
作者: Shuxiao Xie,Shuyang Xie,Yuan Cao,Dezhi Ran,Wei Yang,Tao Xie
类目: Machine Learning (cs.LG); Numerical Analysis (math.NA)
*备注: 25 pages, 4 figures. Reproducibility artifact: this https URL
Abstract:A Strassen-type algorithm has many realizations with the same exact product and multiplication count yet different fp8 error because basis changes reshape coefficient geometry, posing the question of which to run. No current account settles this: classical stability controls worst-case \ell_1 growth, not the expected-error magnitude, and the Dumas–Pernet–Sedoglavic optimizer could only be called probably optimal, its global optimality unproved. To settle this, we attach to each realization a coefficient functional \Phi , a scalar summary of its coefficient geometry, which we minimize over the change-of-basis orbit. This Kempf–Ness problem on a Hadamard manifold lets us certify the global \Phi optimum rather than merely search for it: an exact moment-map zero fixes \Phi_\min = 200/9 , and de Groote’s classification extends that optimality to every exact real rank-7 2\times2 decomposition. Every exact real rank-7 realization therefore has a \Phi -predicted RMS constant at least 5/3 times that of the cubic algorithm, at fixed noise coefficient. We then introduce an explicit block-scaled e4m3 model in which \Phi is the leading-order coefficient of relative expected mean-squared error, and we test the resulting \Phi -predicted ordering against realized fp8 error. Ordering and re-basing experiments support that prediction within tested fused block-scaled regimes, and on real matmul tiles from four architecture families the \Phi -optimal realization falls in the fp8 low-error region. Across two \sim 70B models on real deep_gemm kernels, the same realization removes 10 to 55% of classic Strassen’s excess NLL over the clean model. Algorithm realization thus becomes a mathematically certified design problem rather than a tuning choice: an independent low-precision axis with a global \Phi optimum and measured fp8 relevance.
[LG-30] Differentiable Policy Transport over Multi-Layer Network Feasibility Geometry
链接: https://arxiv.org/abs/2609.26068
作者: Zuyuan Zhang,Zeyu Fang,Mahdi Imani,Nathaniel D. Bastian,Tian Lan
类目: Networking and Internet Architecture (cs.NI); Machine Learning (cs.LG)
*备注:
Abstract:Learning-based control is increasingly central to automating network operations. A learned policy, however, must satisfy cross-layer constraints on interference, power-rate coupling, flow conservation, service chains, capacity, latency, and reliability. Existing methods typically account for only a subset of this geometry and only indirectly, e.g., through reward penalties, Lagrange multipliers, or post-hoc repairs. This paper proposes \emphNetwork Feasibility Geometry Reinforcement Learning (NFG-RL), which models coupled constraints via transport theory and the residual inclusion \bphi_\mathfrakN(x,a)\in\cK_\mathfrakN , defining the executed policy as the pushforward of a proto-policy through a feasibility-transport map. NFG-RL compiles heterogeneous constraints into typed residual blocks and transports proto-actions through a differentiable variational operator, letting active constraints shape execution, exploration, and actor gradients. Our analysis shows that exact transport yields almost-sure feasible execution, while active constraints contract exploration onto the feasible tangent space. It further establishes a nonnegative first-order gain from critic-tilted transport over plain projection and recovers backpressure scheduling as the gradient of a lifted drift residual. In two public-trace-conditioned wireless-edge surrogate environments, NFG-RL improves feasible utility by \textbf37.5–41.5% over the strongest non-NFG method in each environment, reduces raw-action violation by \textbf48.5–60.8%, and lowers P99 delay by \textbf57.0–75.5%, outperforming a range of optimization and learning baselines.
[LG-31] FuncCode: Compressing Kolmogorov–Arnold Networks in Function Space with Hardware-Aware Quantization
链接: https://arxiv.org/abs/2609.26067
作者: Kazi Ahmed Asif Fuad,Lizhong Chen
类目: Machine Learning (cs.LG)
*备注:
Abstract:Kolmogorov–Arnold Networks (KANs) replace scalar edge weights with learnable univariate functions, increasing flexibility but also parameter memory because each edge stores multiple coefficients, often together with a separate base branch. We introduce FuncCode, a basis-agnostic compression approach that forms shared codebooks from sampled edge responses, codes the basis and base branches independently, and exports the resulting codebooks and per-edge indices in a quantized, bit-packed format. Across spline and polynomial KANs, sampled edge responses exhibit 13 – 35% lower effective rank than their coefficient representations. Further replicated controls show that function-space clustering alone is statistically tied with coefficient-space clustering; the consistent accuracy gain comes from preserving the distinct sharing structure of the two branches. On a ten-seed MNIST benchmark, FuncCode compresses spline and GRAM KANs by 31.6\times and 17.6\times with only 0.31 and 0.34 pp accuracy loss. On a 6.1M-edge convolutional KAGN, it achieves 19.9\times compression while remaining within 0.54 pp of dense accuracy on CIFAR-10 and 1.89 pp on CIFAR-100. After compression, per-edge indices account for up to 99.4% of stored weight bits, making the representation index-bound. Across nine bit-exact FPGA accelerators, FuncCode reduces SplineKAN post-route weight memory by 3.87\times relative to dense INT4, without increasing cycle count or latency. The FuncCode implementation is available at this https URL.
[LG-32] owards Adaptive Federated Graph Clustering: A Global Community-aware Contrastive Learning-based Approach
链接: https://arxiv.org/abs/2609.26063
作者: Yinlin Zhu,Di Wu,Wang Luo,Guocong Quan,Miao Hu
类目: Machine Learning (cs.LG)
*备注: Under Review
Abstract:Federated graph learning (FGL) enables multiple clients to collaboratively train graph models without sharing their private graph data, providing a promising paradigm for mining knowledge from distributed graph repositories. While most existing FGL methods focus on supervised tasks, real-world graphs are often massive and unlabeled, making federated graph clustering an important yet still immature research direction. Notably, this task is particularly challenging due to the inherent subgraph heterogeneity across clients, which leads to client-specific community structures. In this work, we identify two critical limitations in existing federated graph clustering methods: (1) unrealistic pre-defined cluster cardinality assumptions and (2) incomplete inter-community separation. To address these challenges, we propose AdaFGC, an Adaptive Federated graph clustering framework based on Global community-aware Contrastive learning. AdaFGC introduces an over-complete set of global community anchors to model the global community structure and adaptively estimate clustering cardinality via cross-client anchor refinement. In addition, it employs a global community-aware contrastive learning scheme that uses the shared anchors as contrastive prototypes to explicitly enforce community-level attraction and repulsion across clients, complemented by node-level and topology-level objectives that stabilize local representations. Extensive experiments on eight benchmark datasets demonstrate that AdaFGC consistently outperforms existing supervised and unsupervised FGL baselines across multiple clustering metrics.
[LG-33] BOBA: Dynamic Bayesian Optimization through Bayesian Active Inference
链接: https://arxiv.org/abs/2609.26021
作者: Merlin Angel Kelly,Rishan Patel,Alexander Thomas,Ziyue Zhu,Zikun Quan,Tom Carlson,Youngjun Cho
类目: Machine Learning (cs.LG)
*备注:
Abstract:Dynamic black-box optimization presents significant challenges for Bayesian Optimization (BO), as the objective function evolves over time, causing optimal locations to shift continuously. Existing dynamic BO (DBO) methods using standard acquisition functions such as Upper Confidence Bound (UCB) fail to explicitly account for temporal variations, leading to suboptimal sample allocation and poor tracking of moving optima. Here, we propose BOBA (Bayesian Optimization through Bayesian Active Inference), a novel acquisition function inspired by free energy principles from active inference that explicitly minimizes predictive uncertainty about future states in dynamic environments. BOBA extends traditional acquisition functions by incorporating a forward-looking uncertainty quantification that estimates uncertainty in function changes, enabling more informed exploration-exploitation trade-offs in non-stationary settings. We evaluate BOBA on synthetic dynamic benchmarks, comparing against state-of-the-art DBO methods. Our experiments demonstrate that BOBA significantly improves regret in query-restricted settings, while remaining competitive in time-limited settings. We further analyze variants of BOBA with different exploration strategies, showing how the exploration-exploitation balance can be tuned for different types of dynamic functions. This work contributes both a free energy-based acquisition function for DBO and insights into how active inference principles can enhance optimization in non-stationary environments, with implications for real-time applications requiring continuous adaptation.
[LG-34] he Dynamics of Quasiregular Neural Learning
链接: https://arxiv.org/abs/2609.26018
作者: Matthia Sabatelli
类目: Machine Learning (cs.LG)
*备注:
Abstract:Many learning problems combine a dominant regularity with systematic exceptions. Motivated by U-shaped learning in language acquisition, we study this interaction in controlled quasiregular regression problems where regular and exceptional solutions are explicitly known. Neural networks can partially acquire exceptions, subsequently regress toward the dominant regularity, and finally recover. This overregularization becomes substantially stronger when exceptions are rare, despite their early acquisition, but does not emerge equally across all regularities considered. Our results isolate a simple form of competition between regularities and exceptions during neural learning.
[LG-35] heory for groupoid equivariant neural networks: an approach for steerable CNNs on bounded domains
链接: https://arxiv.org/abs/2609.25987
作者: Alberto Ibort,Maria Jimenez-Vazquez,Juan M. Perez-Pardo
类目: Machine Learning (cs.LG); Differential Geometry (math.DG)
*备注:
Abstract:Equivariant convolutional neural networks are usually built from a group acting globally on the space of signals. This hypothesis is inappropriate for many bounded or stratified domains: an ambient rigid motion may be admissible only on part of the domain, and the boundary introduces geometric types that are invisible to a transitive group action. We develop a theory of groupoid-equivariant neural networks in which the symmetry datum consists of a groupoid, a selected pseudogroup of local bisections, a measure, and input and output representation bundles. For integral channels on the object space, we prove a bisection-equivariant kernel theorem: equivariance is equivalent to a transport constraint on the two-point kernel, and its solutions are classified by one joint-stabilizer intertwiner on each orbit of pairs. As a case study we apply the theory to bounded planar domains. The resulting architecture is implemented through offline nullspace bases and sparse gather–transform–scatter operations. A Poisson–Dirichlet kernel study is used separately to assess boundary-aware inductive bias; the exact inverse is shown to preserve the global symmetries of the rectangle but not general proper local bisections. The numerical results show that the proposed architectures provide significant advantages when symmetries cannot be globally implemented by group actions and provide an accuracy improvement of at least one order of magnitude with respect to the models tested. Subjects: Machine Learning (cs.LG); Differential Geometry (math.DG) MSC classes: 53Z50, 68T07 ACMclasses: I.2.6 Cite as: arXiv:2609.25987 [cs.LG] (or arXiv:2609.25987v1 [cs.LG] for this version) https://doi.org/10.48550/arXiv.2609.25987 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
[LG-36] GeoPair: Geometry-Preserving Cross-Layer Factorization for Training-Free Transformer Compression
链接: https://arxiv.org/abs/2609.25963
作者: Baher Mohammad,Ammar Ali,Stamatios Lefkimmiatis
类目: Machine Learning (cs.LG)
*备注:
Abstract:Transformer architectures exhibit cross-layer redundancies, yet post-training compression pipelines typically optimize layers in isolation or rely on heuristic grouping strategies that disregard layer-specific activation geometries. We introduce a principled, training-free framework that sequentially optimizes cross-layer weight pairings and shared-dictionary factorizations. Rather than forcing weights of adjacent layers to share a basis or heuristically merging activation statistics, our approach identifies structurally compatible projections and learns a shared representation that better preserves each layer’s distinct calibration geometry. Coupled with structured sparsity, this yields highly efficient weight decompositions without sacrificing functional fidelity. Across diverse architectures, scales, and modalities, our method achieves state-of-the-art results, consistently outperforming independent structured weight decompositions and alternative pairwise weight factorizations, which operate under heuristic grouping strategies. By replacing heuristic engineering strategies with a convergent, optimization-driven pipeline, we establish a theoretically grounded foundation for scalable, transformer compression across different modalities.
[LG-37] Exploring Solver-Level Warmstarting for Neural Network Verification
链接: https://arxiv.org/abs/2609.25962
作者: Annelot Bosman,Minghao Liu,Marta Kwiatkowska,Holger Hoos,Jan van Rijn
类目: Machine Learning (cs.LG)
*备注: to be published in the postproceedings of WORKSHOP ON SECURE AND TRUSTWORTHY AI (2026) co-located with the European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases
Abstract:Neural network verification has become a key tool for providing formal guarantees on the behaviour of neural networks. However, many verification problems remain computationally intractable in the worst case: even for common adversarial robustness specifications, verification is NP-complete. Here, we explore the application of solver-level warmstarting for neural network verification to exploit information from previous solutions. We study the effect on running time as several properties are modified, including perturbation radii, input data and the networks themselves, using a pipeline that is generalisable and potentially adaptable to state-of-the-art verifiers. Our results show that warmstarting can significantly reduce verification time in most cases. Moreover, warmstarting enables the successful verification of instances that could not be solved from scratch within the given time limit.
[LG-38] Beyond Scalar Sensitivity: Activation-Aware Mixed-Precision LLM Quantization with Cross-Layer Refinement
链接: https://arxiv.org/abs/2609.25916
作者: Akihiro Yoshida,Yuma Ichikawa
类目: Machine Learning (cs.LG); Applications (stat.AP); Methodology (stat.ME); Machine Learning (stat.ML)
*备注: 33 pages, 7 figures
Abstract:Mixed-precision weight quantization is commonly formulated as a Multiple-Choice Knapsack Problem (MCKP), yet existing solvers rely on scalar sensitivity proxies that collapse each weight matrix’s Hessian into a single number and treat every module independently. We prove that even the optimal scalar proxy incurs multiplicative distortion up to \sqrt\kappa(\mathbfA)\kappa(\mathbfB) relative to the full activation-aware quadratic, where \kappa(\mathbfA) and \kappa(\mathbfB) denote the condition numbers of the input- and output-side Hessian factors. This bound varies from 10^1 to 10^13 for typical LLM modules, making inter-module sensitivity ranking unreliable. To address these limitations, we propose Cross-layer Activation-aware Sensitivity Allocation (CASA), a two-phase method. In Stage 1, the scalar proxy is replaced by an activation-aware metric derived from the Kronecker-factored Hessian, reducing the MCKP to a form whose continuous relaxation admits a closed-form solution. In Stage 2, a cross-layer-aware local search evaluates bit-width updates using the end-to-end model loss. Experiments on multiple LLMs across different bit budgets show that CASA achieves lower perplexity than the latest scalar-proxy baselines, especially at ultra-low bit-widths ( 3 bits per weight). Moreover, the performance gain in zero-shot accuracy tracks the per-model average condition-number over modules, confirming the distortion bound as a practical indicator of scalar-proxy failure.
[LG-39] AURA: Angular Update Rate Adaptation for training complex-valued neural networks
链接: https://arxiv.org/abs/2609.25914
作者: Enrico Ballini,Allan Peter Engsig-Karup,Tito Andriollo
类目: Machine Learning (cs.LG); Numerical Analysis (math.NA)
*备注:
Abstract:Complex-valued neural networks (CVNNs) are increasingly adopted for complex-valued data; however, they are often trained with first-order optimizers inherited from the real-valued case. The efficiency of these methods depends largely on the step size, and their step-size rules ignore the angular information available in the complex plane. We address step-size adaptation in the complex domain by introducing AURA (Angular Update Rate Adaptation), a per-parameter step-size adaptation that can be added on top of any first-order optimizer, and removed from it, without altering its update direction. AURA measures the agreement between consecutive updates of each complex parameter, in length, alignment, and sense of rotation, and enlarges the step when they are consistent and reduces it when they are not. It requires no additional gradient evaluations and only inexpensive vector operations per step. We combine AURA with Adam and Muon and compare the resulting methods with well-known first-order optimizers on four test cases of increasing complexity, ranging from the approximation of scalar complex functions to physics-informed training. Fully connected neural networks are used throughout this work. All hyperparameters other than the step size are held fixed across test cases; for one case, we also tune the hyperparameters of each optimizer under the same budget. Our empirical tests show that AURA improves the convergence of its base optimizer in most cases with a small per-step overhead, and we identify the conditions under which it fails to do so.
[LG-40] Neural Approximation by Function Composition: Rigidity and Doubly Exponential Convergence
链接: https://arxiv.org/abs/2609.25874
作者: Wentao Huang,Haizhang Zhang
类目: Machine Learning (cs.LG); Information Theory (cs.IT)
*备注:
Abstract:Deep neural networks approximate functions by composing affine maps with nonlinear activations, but how composition itself creates approximation power is not yet fully understood. We investigate a fundamental mechanism: geometrically weighted sums of iterates of a single scalar generator function. This mechanism underpins the classical tent-map construction of the function (x - x^2) and related recursive representations used by Yarotsky, W. E, et al., to analyze the approximation powers of deep neural networks. First, we establish a rigidity theorem: for continuous piecewise linear generators with a finite number of segments, any (C^3) function that can be represented in this way is at most quadratic. For non-affine quadratic functions, the geometric factor is at least 1/4 . This result both reveals limitations of the tent-map approach and complements existing methods based on hierarchical bases and recursive polynomial constructions. Second, using an exact remainder identity as guidance, we construct a smooth generator whose iterates yield doubly exponential error decay in total depth for square approximation and, through multiplication modules, for each fixed polynomial. For power series with absolutely summable coefficients on ([-1,1]^d), distributing depth according to monomial degree yields a uniform approximation error of order (O(e^-cL^1/d)) on each interior cube. These findings demonstrate how generator dynamics and remainder estimates govern depth allocation and approximation rates of deep neural networks. Subjects: Machine Learning (cs.LG); Information Theory (cs.IT) Cite as: arXiv:2609.25874 [cs.LG] (or arXiv:2609.25874v1 [cs.LG] for this version) https://doi.org/10.48550/arXiv.2609.25874 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
[LG-41] Gaussian Flow-Matching Schedules: Implications for Sampling and Training
链接: https://arxiv.org/abs/2609.25839
作者: Arsène Claustre(DI-ENS),Hugo Negrel(DMA, CFM),Claire Boyer(LMO, IUF),Kimia Nadjahi(DI-ENS),Eric Vanden-Eijnden(DMA, CFM, CIMS)
类目: Machine Learning (cs.LG)
*备注:
Abstract:Flow-matching schedules affect both sampling dynamics and the variance of the regression target. For centered commuting Gaussians, we show that a direction-dependent schedule decomposes into two independent design choices: a variance path, which fully determines the intermediate laws and probability flow, and a factorization, which leaves this flow unchanged while controlling irreducible regression variance. On the sampling side, we analyze finite-step Euler accuracy and derive a necessary drift bound for exact N -step sampling, connecting the geodesic and the logarithmic path. On the training side, for any fixed path, we derive closed-form factorizations that either minimize time-averaged regression variance or make it constant along the path.
[LG-42] Protocol before progress: leakage-aware evaluation of AIS trajectory prediction
链接: https://arxiv.org/abs/2609.25827
作者: Zobeir Raisi,Vali Mohammad Nazarzehi Had
类目: Machine Learning (cs.LG)
*备注:
Abstract:Reported gains in vessel-trajectory prediction from Automatic Identification System (AIS) data are credited to new architectures, but the evaluation protocol is rarely measured as a source of error reduction. We build a leakage-aware protocol with vessel-, time- and region-disjoint splits and apply it to two corpora with different traffic: 31 days of Danish national AIS traffic and 30 days of US Gulf coast traffic off Houston and Galveston. On both, we audit TrAISformer, GATransformer, and controlled AISFormer-inspired reconstructions. Three protocol effects appear in both corpora. First, TrAISformer’s best-of-16 oracle decoder lowers error by a factor of 2.1-3.2 relative to greedy decoding. Second, a split that shares vessels lowers its greedy error by 23-25% at one hour, against 2% or less for a compact 0.43 M-parameter encoder. Third, a region-disjoint split raises TrAISformer’s one-hour error from 2.2 to 24.6 km on the US corpus, because 99.9% of the test contexts fall in longitude bins never seen in training; the encoder built on local offsets is unaffected by this. Architectural mechanisms matter less: GATransformer’s graph attention gives no measurable benefit on either corpus, while its waterway feature is worth 12-22%. The effect of a time-disjoint split is not stable across corpora (13% versus 2%). We release the splits and code.
[LG-43] Beyond Reconstruction Error: Analytical and Data-Driven Action Tokenization for Autoregressive Vision-Language-Action Models
链接: https://arxiv.org/abs/2609.25820
作者: Yuxin Yang,Gaohan He,Changxue Guan,Hangming Liu
类目: Robotics (cs.RO); Machine Learning (cs.LG)
*备注: 5 pages, 1 figure, 4 tables
Abstract:Discrete action tokenization is central to autoregressive vision-language-action (VLA) models, yet action representations are often evaluated primarily through reconstruction fidelity. We ask which representation properties actually matter for closed-loop control by comparing fixed analytical, data-driven linear, and nonlinear neural representations under a unified tokenization interface. Across rate-distortion analysis, sequence-modeling diagnostics, and 3,500 LIBERO rollouts, representation rankings change with the evaluation criterion. PCA achieves lower nominal reconstruction error than Temporal-DCT, but produces less predictable token sequences and 3.0 percentage points lower mean seen-task success across three policy-training seeds, with the policy ordering reversing in one seed. In a matched seed-42 ablation, an autoencoder further reduces reconstruction error yet does not yield the strongest policy and exhibits greater sensitivity to discrete token perturbations. These findings show that reconstruction fidelity alone cannot reliably select action representations for autoregressive control, motivating joint evaluation of geometric fidelity, sequence predictability, decoder stability, and closed-loop performance.
[LG-44] CacheDyG: Decoupling Temporal Propagation for Efficient Dynamic Graph Learning
链接: https://arxiv.org/abs/2609.25814
作者: PinHeng Zong,Ye Yuan
类目: Machine Learning (cs.LG)
*备注: Accepted at ADMA 2026. 17 pages, 3 figures
Abstract:Dynamic graphs are widely used to model time-evolving relational systems in real-world applications. Dynamic graph neural networks provide an effective framework for capturing both structural dependencies and temporal dynamics in such data. However, they typically intertwine temporal graph propagation with every optimization epoch and often maintain large trainable representations for each node-time pair. This design repeatedly recomputes largely unchanged historical structures, leading to substantial training and parameter overhead. To address this critical issue, we propose CacheDyG, a Cache-refine framework for efficient Dynamic Graph learning. Specifically, it decouples temporal propagation from routine parameter updates by constructing a time-ordered temporal dependency cache that stores graph-aware node-time representations in non-trainable buffers. During standard training epochs, CacheDyG reads from the cache and updates only a lightweight cache refiner, an adaptive residual gate, and the link predictor. Selective cache refresh further keeps cached representations aligned with the supervised objective while avoiding epoch-wise sparse propagation. Experiments on five dynamic graph benchmarks show that CacheDyG adopts substantially fewer trainable parameters and lower runtime to obtain more competitive predictive performance than baselines. These results demonstrate that cache-based decoupling provides an effective principle for scalable dynamic graph learning.
[LG-45] Multi-View Fair Clustering Guided by Cross-View Sensitive Information Discrepancy
链接: https://arxiv.org/abs/2609.25811
作者: Mudi Jiang,Jiahui Zhou,Xinying Liu,Zengyou He,Zhikui Chen
类目: Machine Learning (cs.LG)
*备注:
Abstract:Multi-view clustering (MVC) aims to uncover latent cluster structures by exploiting complementary information from multiple views. Despite substantial progress in clustering performance, fairness remains an important concern when MVC is applied to socially sensitive scenarios. Recent fair multi-view clustering methods have introduced fairness constraints into representation learning or clustering assignments. However, these methods generally treat different views under a largely uniform fairness mechanism, without explicitly distinguishing their varying levels of sensitive dependence during cross-view learning. In practice, different views may encode substantially different levels of sensitive information. Ignoring such cross-view discrepancy can allow highly sensitive-dependent views to influence less sensitive-dependent ones during cross-view learning, potentially degrading both clustering performance and fairness. To address this issue, we propose a novel multi-view fair clustering framework guided by cross-view sensitive information discrepancy. Specifically, we estimate the sensitive dependence of each view and develop a bias-ranked asymmetric alignment mechanism that encourages views with higher sensitive dependence to learn from those with lower sensitive dependence, while cross-view discrepancies are further exploited to adaptively regulate the alignment process. Moreover, fairness regularization is imposed on the consensus soft assignments to further promote group fairness. Extensive experiments on benchmark datasets demonstrate that the proposed method achieves a favorable balance between clustering quality and group fairness.
[LG-46] A Lightweight Plastic-Memory Framework for Graph Few-Shot Class-Incremental Learning
链接: https://arxiv.org/abs/2609.25781
作者: Zihan Mei,Zhili Qin,Tongze Zhang,Hongyuan Liu,Junming Shao,Qinli Yang
类目: Machine Learning (cs.LG)
*备注: 9 pages, 3 figures
Abstract:Graph Incremental Learning has garnered increasing attention as dynamic graph data continues to emerge across diverse fields. Conventional approaches primarily address catastrophic forgetting by preserving node-related knowledge through replay or distillation techniques; however, they often incur high computational costs and inefficiency. This issue is further exacerbated in real-world scenarios where labeled data for new classes is scarce. In this paper, we propose a novel lightweight plastic-memory framework specifically designed for few-shot incremental learning on graphs. The core idea of our framework is the construction of a plastic-memory module that evolves over time, continuously updating and expanding its memory to accommodate new classes while retaining previously learned knowledge. In contrast to existing techniques, our memory module is both lightweight and effective, featuring an innovative evolving micro-clustering structure that dynamically updates representations of class prototypes, sub-prototypes, and their interaction weights. Building on this memory module, we introduce a memory-driven meta-learning framework that enhances adaptability to new tasks in its inner loop while maintaining stability for earlier tasks in the outer loop. Extensive experiments on four benchmark datasets demonstrate the framework’s superior performance in balancing stability for old knowledge and adaptability to new knowledge.
[LG-47] Disentangling Heterogeneous Traffic Dynamics for Multi-Step Traffic Forecasting via Adaptive Spectral Decomposition
链接: https://arxiv.org/abs/2609.25777
作者: Zijun Huang,Chenrui Fu,Wenhao Wang,Xiaochuan Gou,Chih-Chieh Hung,Guanyao Li
类目: Machine Learning (cs.LG)
*备注: 13 pages, 1 figure, 3 tables
Abstract:Accurate multi-step traffic forecasting remains challenging because observed traffic signals contain heterogeneous temporal dynamics with different characteristics and levels of predictability. Existing approaches typically model these dynamics within a unified representation or rely on predefined decomposition rules, which may limit their ability to flexibly separate persistent patterns from rapidly varying fluctuations. To address this issue, we propose the Adaptive Decomposition Network (ADNet), a component-specific forecasting framework that adaptively disentangles traffic dynamics into dominant and residual components. ADNet introduces a learnable complementary spectral decomposition mechanism that determines the contribution of each frequency bin to the two components. Unlike hard frequency partitioning, every frequency bin can contribute to both components with different learned proportions, allowing the decomposition to be optimized jointly with the forecasting objective. The reconstructed components are then modeled by two dedicated spatiotemporal forecasting branches, and their predictions are integrated to generate the final multi-step forecast. Experiments on the Alameda and Orange regions of the TraffiDent dataset show that ADNet achieves the best performance in 20 of the 24 reported region-horizon-metric comparisons, with particularly clear gains at longer forecasting horizons. Capacity-controlled ablation experiments further show that the learnable decomposition substantially outperforms a fixed decomposition and provides additional improvements beyond the dual-branch architecture alone. These results demonstrate the effectiveness of adaptive decomposition and component-specific modeling for multi-step traffic forecasting.
[LG-48] Minimal Recurrent Behavioral Memory for Imitation under Partial Observability
链接: https://arxiv.org/abs/2609.25757
作者: Xianyao Li,Fang Xu,Rui Min,Ruitong Tian,Jing Du
类目: Machine Learning (cs.LG); Information Theory (cs.IT); Robotics (cs.RO)
*备注: 46 pages, 10 figures. Code: this https URL
Abstract:What is the least recurrent memory needed to reproduce a specified expert under partial observability? The instantaneous requirement is the conditional entropy of the expert’s behavioral quotient, but recurrence must also preserve distinctions that future observations will not restore before use. We characterize this minimal recurrent behavioral memory by a compatibility relation: under transitivity its classes attain the exact minimum, while the general case is an entropy minimization over closed compatible state assignments, with exact certificates on finite instances. A sole-carrier measurement protocol separates behavioral sufficiency, excess code rate, and information carried by observations or other memory paths; experimental bit requirements refer to the induced symbolic behavioral model under the stated occupancy. Across manipulation tasks, learned code rates remain near zero- and two-bit requirements as hidden modes grow to 512 , and anticipatory memory follows a 2\to1\to0 requirement despite zero instantaneous demand during waiting. Learning this representation remains difficult: event-agnostic future-behavior supervision yields 36/40 sufficient seeds with one frozen configuration and improves the longest-horizon pixel setting from 0/8 to 6/8 sufficient held-out seeds (closed-loop success from 0.08 to 0.57 ). On unmodified community benchmarks, the protocol certifies delay-independent requirements, which sufficient codes match at mid-delay. The supervision aids commitment but can induce predictive surplus; annealing it lets imitation and rate training reduce that surplus, separating the information-theoretic target from the ability to learn it.
[LG-49] Modular Norm RandOpt: Population-Efficient Ensembling through Architecture-Aware Perturbations
链接: https://arxiv.org/abs/2609.25745
作者: Kirato Yoshihara,Hiroaki Hamade
类目: Machine Learning (cs.LG)
*备注: Preprint. Project page: this https URL
Abstract:RandOpt samples weight-perturbed language models and ensembles top-ranked candidates through plurality voting, but its global perturbation scale ignores heterogeneous module geometry. We propose \mbox\textbf\emphModular Norm RandOpt, an architecture-aware sampling method using module-wise natural norms and calibrated scales while preserving selection and voting. It outperforms RandOpt using 3\times fewer candidates on Countdown and at least 12\times fewer on GSM8K, with corresponding wall-clock savings. Evaluations across seven tasks and three Qwen scales ( 0.5 B-- 3 B) show higher mean accuracy than RandOpt on Countdown, GSM8K, and MATH-500 at every scale. The gains extend to Llama 3.2 3 B and Gemma 3 4 B on Countdown and GSM8K. On Qwen2.5-1.5B, our ensembles also achieve higher mean accuracy than iterative baselines on both tasks at comparable main-run evaluation budgets. On GSM8K, a tail-density diagnostic implies only a 1.2 – 1.8\times candidate reduction, while most ensemble improvement is associated with more favorable correct-expert support. These results highlight perturbation geometry as a key design choice for population-efficient, gradient-free search around pretrained models.
[LG-50] Signed Graph Pre-Training and Prompt Learning
链接: https://arxiv.org/abs/2609.25722
作者: Zihan Mei,Rong Pan,Yuzhou Chen,Yixuan He
类目: Machine Learning (cs.LG)
*备注: 25 pages, 3 figures, Accepted to Learning on Graphs Conference (LoG 2026)
Abstract:Signed graphs arise in trust–distrust networks, financial correlation systems, biological interaction graphs, and many other domains in which edges can be positive or negative and may also be directed. While signed graph neural networks have improved task-specific learning, graph transfer learning on signed graphs remains underdeveloped. In this paper, we introduce TopoSIGN, a pioneer topology-guided graph pre-training and prompt learning framework for signed graphs. TopoSIGN combines a structural encoder built on the magnetic signed Laplacian with a novel persistent-homology branch that summarizes signed topology through Dowker-complex persistence images. The fused embeddings are then transferred to a prompt learning function. Experimental results on synthetic and real-world datasets demonstrate the efficacy of TopoSIGN in extracting useful structural information in signed graphs, as well as the adaptability and flexibility of the proposed general framework.
[LG-51] Graph Domain Adaptation Does Not End with Representation Learning
链接: https://arxiv.org/abs/2609.25692
作者: Ziqian Liu,Yongxue Xu,Enze Zhang,Jiaqi Zhang,Hao Wang,Maolin Wang
类目: Machine Learning (cs.LG)
*备注:
Abstract:Graph domain adaptation (GDA) transfers knowledge from a labeled source graph to an unlabeled target graph under shifts in both node attributes and graph structure. Existing methods primarily adapt graph representations through propagation redesign, distribution alignment, or source-to-target transition modeling, but still rely on a single graph-propagating path for target prediction. This leaves open whether an adapted graph representation exhausts the predictive evidence available in the target domain, since the graph-aware expert and graph-free local expert may exhibit different failure modes under topological shifts. To address this limitation, we propose EviGDA, an Evidence-Augmented Graph Domain Adaptation framework that complements graph representation adaptation with a graph-free local expert. The graph-aware expert performs message passing and entropy-aware marginal alignment, while the graph-free local expert learns solely from source node features and labels without graph propagation or target alignment. The two experts are optimized independently and combined only at inference through a task-level constant probability mixture, preserving complementary evidence without joint training, learned routing, or target pseudo-labels. Extensive experiments on ten datasets and 16 transfer tasks show that EviGDA outperforms state-of-the-art baselines.
[LG-52] Marginal Log-Likelihood Increments under Dirichlet-Smoothed Markov Estimation
链接: https://arxiv.org/abs/2609.25675
作者: Levin David Schwab
类目: Machine Learning (cs.LG)
*备注: 14 pages, 1 figure, 3 tables
Abstract:For a Dirichlet-smoothed transition model, the effect of adding one workflow trace to the training archive is an exact change in reference-weighted log likelihood. We derive that change and show that it is a weighted reduction of Kullback–Leibler divergence between the reference conditionals and the model. From this form we obtain an upper bound on the gain available to any acquisition, which expresses a millinat difference as a share of what is attainable, an exact covariance identity for the effect of the reference weighting, and a sign criterion for the interaction between two candidates, from which the batch objective is neither submodular nor supermodular. A case study on the BPI Challenge 2012 loan-application log measures all three and finds a positive selection result in one of the four combinations of reference weighting and budget unit. There, of two regressors fitted to identical descriptors and identical labels, the one that predicts individual increments far more accurately, median R^2 0.87 against 0.62, realizes the smaller share of the attainable gain, 61 against 69 per cent, so ranking accuracy for individual traces is neither necessary nor sufficient for batch quality.
[LG-53] When Riemann flows with Wasserstein: Generative Modeling of Probability Distributions on Manifolds
链接: https://arxiv.org/abs/2609.25659
作者: Doron Haviv,Edward De Brouwer,Rishabh Anand,Rex Ying,Aïcha Bentaieb,Gabriele Scalia,Hector Corrada Bravo
类目: Machine Learning (cs.LG)
*备注:
Abstract:Many scientific datasets, such as molecular conformational ensembles or single-cell tissue measurements, are naturally modeled as meta-distributions: distributions over probability measures on non-Euclidean domains. Existing generative methods largely assume Euclidean geometry and fail to capture this structure. We introduce Riemannian Wasserstein Entropic Flow Matching (RWEFM), a generative framework on the Wasserstein space \mathcalP_2(\mathcalM) of a Riemannian manifold (\mathcalM,g) . RWEFM is trained by regressing a neural vector field onto Riemannian optimal transport velocities, using McCann displacement interpolations as conditional paths. We confirm theoretically that this construction leads to a valid flow matching approach on \mathcalP_2(\mathcalM) and introduce the Riemannian Entropic Map, a GPU-efficient approximation of the optimal transport map on manifolds. Our experiments show that by respecting the intrinsic geometry of the data, RWEFM can generate whole single-cell samples in hyperspherical latent spaces and protein conformational ensembles on the torus. As RWEFM requires only a geodesic distance and a projection operator, it is not restricted to manifolds with closed-form geometry, which we demonstrate by generating distributions on a general triangulated mesh.
[LG-54] argeted Review for AI-Assisted Biodiversity Surveys: Active Continuous-Score Occupancy Modeling
链接: https://arxiv.org/abs/2609.25657
作者: Timm Haucke,Lauren Harrell,Justin Kay,Mary Clapp,Sara Beery
类目: Machine Learning (cs.LG); Computational Engineering, Finance, and Science (cs.CE)
*备注:
Abstract:We increasingly use machine learning to label scientific datasets. The models we develop and deploy are improving all the time, but they are not and will likely never be perfect. Mistakes matter, as errors can propagate into our scientific understanding, particularly when systematically biased. Very reasonably, scientists thus review substantial proportions of ML-generated labels to verify or correct mistakes in pursuit of ensuring their scientific findings are not biased by ML. In this work, we focus on helping scientists optimally allocate this reviewing effort relative to their scientific goals. We focus on a specific class of scientists (ecologists) and a specific, widespread, and impactful modeling target (occupancy modeling, which estimates where species are likely to occur, conditioned on environmental factors). We introduce Active Continuous-Score Occupancy Modeling (ACORN), a method that incorporates ML predictions into occupancy models and strategically selects samples for expert review that are maximally informative for downstream ecological analysis. Across camera-trap and bioacoustic datasets, our method recovers ecological conclusions close to those obtained from fully human-labeled data, while requiring substantially fewer expert reviews than non-targeted review policies. Our results suggest that ML-assisted scientific workflows should optimize expert effort for downstream inference, rather than for classifier accuracy alone, especially when human review budget is limited. Our code is available at this https URL
[LG-55] Accelerating the Mitigation of LLM Inference Nondeterminism Across GPU Architectures
链接: https://arxiv.org/abs/2609.25624
作者: Liam Cooper,Shinnung Jeong,Hyeran Jeon,Jeffrey Young,Hyesoon Kim
类目: Hardware Architecture (cs.AR); Machine Learning (cs.LG)
*备注: 14 pages, 5 figures
Abstract:Large language model (LLM) outputs are expected to be reproducible under greedy decoding, yet in practice the same model, prompt, and software stack produce different outputs on different GPUs. The root cause is floating-point non-associativity combined with hardware-dependent kernel selection. Inference frameworks select different matrix-multiplication kernels on each architecture, with different parallel reduction orders and unspecified tensor-core arithmetic, and the resulting rounding differences can flip output tokens. Existing solutions have imperfect cross-architecture reproducibility and incur a significant performance penalty. We present a solution employing a set of fixed-configuration fused-upcast GEMM kernels that load 16-bit weights from memory, upcast them to FP32 in registers, and accumulate with IEEE-754 arithmetic in a reduction order that is a pure function of the problem shape and is therefore independent of the device, its SM count, or kernel scheduling. By fixing the floating-point reduction order as a function of problem shape alone, every GPU runs the same operation sequence, so cross-architecture reproducibility of the linear layers reduces to correct IEEE-754 arithmetic rather than to rounding differences staying below a tie-flip threshold. We confirm our solution’s linear-layer outputs are bitwise identical across NVIDIA Ampere, Ada, and Hopper GPUs, while running 1.17 to 3.1\times faster end-to-end than the state-of-the-art solution and cutting weight-memory traffic in half.
[LG-56] SambaGraph: Action-Reaction Spatio-Temporal Graphs for Soccer Tactical Response Modeling
链接: https://arxiv.org/abs/2609.25569
作者: Abel A. Reyes-Angulo,Henry O. Velesaca,Steven Araujo
类目: Machine Learning (cs.LG)
*备注:
Abstract:Soccer tactics are interactive: an attacking action changes the opponent’s defensive problem, and the observed response depends on the multi-agent match state. We introduce SambaGraph, an action–reaction spatio-temporal graph dataset and benchmark for soccer tactical response modeling. From tracking and event data for all 64 matches of the 2022 FIFA World Cup, we curate 4,070 action-centered episodes represented as temporally aligned 23-node player–ball graph sequences with attack/defense views, response labels, and 26,270 split-safe attack–defense pairs. We study three questions: whether observed responses can be classified from graph episodes, whether successful defenses can be retrieved for a query attack, and whether graph-derived summaries support grounded LLM reasoning. A compact signature MLP obtains 0.796\pm0.007 macro-F1 for response classification, while a fused graph–signature dual encoder reaches 0.471\pm0.029 Hit@5 and 0.655\pm0.051 Hit@10 for full-bank defensive retrieval. Hard negatives maximize pair discrimination but not retrieval quality. Local LLMs underperform supervised encoders for direct classification and do not improve over a strong original order in eight-candidate reranking, but they provide grounded tactical rationales. These results position SambaGraph as a reproducible benchmark for graph-based soccer strategy-response research. Code and dataset are available at: this https URL.
[LG-57] HABILIS Brain 0: Geometry-Change Supervision for Vision-Language-Action and Residual Flow Recovery
链接: https://arxiv.org/abs/2609.25558
作者: Jinu Pahk,Jesoon Kang,Taegeon Park,Jisu An,Soo Min Kimm,Jaejoon Kim,Byoung-Tak Zhang
类目: Robotics (cs.RO); Machine Learning (cs.LG)
*备注:
Abstract:Vision-language-action policies benefit from geometric supervision, but current-frame geometry alone does not explicitly describe the changes associated with manipulation. This design is motivated by the goal of learning an embodiment-agnostic visual interface that can be pretrained across robot and egocentric video before robot-specific action alignment. We introduce Geometry-Change VLA (GC-VLA), which learns to predict multiview future-current geometry-change tokens from current observations. Offline frame pairs define a nominal 0.5-second prediction horizon; future observations are used only to construct training targets. Stage 1 trains a geometry-change vision-language model (GC-VLM). Stage 2 introduces a continuous ActionExpert and aligns it with robot actions while stopping action-flow gradients at the VLM interface. Stage 3 enables these gradients to update the trainable VLM components jointly with the ActionExpert. Stage 4 freezes GC-VLA and applies Geometry-Conditioned Residual Flow (GCRF), using a binary intervention router and a single bounded residual velocity policy learned from closed-loop feedback. GC-VLA achieves 95.20% success on LIBERO, and GC-VLA with GCRF achieves 99.55%. Inference uses current observations and the learned GC representation without executing the offline target encoders.
[LG-58] Synthesis and editing of multi-instrument audio mixtures using scalar-quantised latents with MIDI Span conditioning ICASSP2027
链接: https://arxiv.org/abs/2609.25546
作者: Sungkyun Chang,Keshav Bhandari,Simon Dixon,Emmanouil Benetos
类目: ound (cs.SD); Machine Learning (cs.LG); Audio and Speech Processing (eess.AS)
*备注: 13 pages including appendices, 2 figures. Submitted to ICASSP 2027
Abstract:Music creation often involves iterative refinement, changing selected musical details while retaining the rest. To support such refinement, we introduce SpanSynth-Edit, a flow-matching model for MIDI-guided synthesis and editing of multi-instrument audio mixtures using low-frame-rate scalar-quantised latents. MIDI Span encodes instrument-labelled note lifecycles as unordered event sets with continuous-valued attributes and pools each set into one conditioning vector per audio-latent frame. The model uses contextual audio for instrument-specific timbre guidance and supports editing by resynthesising the target region from revised MIDI. Experiments on single- and multi-instrument benchmarks show competitive performance and demonstrate within-frame onset control. We also discuss limitations of transcription-based note-adherence evaluation.
[LG-59] Continuous Optimization for p-adic Models
链接: https://arxiv.org/abs/2609.25501
作者: Julian Salazar,Dimitri Kanevsky,Matt Harvey,Pascal Getreuer,Lucas Dixon
类目: Machine Learning (cs.LG); Metric Geometry (math.MG); Number Theory (math.NT); Optimization and Control (math.OC); Machine Learning (stat.ML)
*备注:
Abstract:We present the first method for native, continuous gradient descent for machine learning models with p -adic parameters. Existing native optimizers are discrete, mostly combinatorial searches, as the p -adic numbers \mathbbQ_p are totally disconnected, with standard losses that are flat away from their minima. To enable continuous optimization, we propose working with \mathbbQ_p via its Berkovich affine line: a canonical, path-connected expansion of \mathbbQ_p that preserves its isometries and uniquely extends its analytic maps. This hull is a metric tree with interpretable points and local derivatives, which we show enables effective optimizers and backpropagation. We formulate gradient descent and show that its approximations efficiently learn linear models with coefficients in \mathbbQ_p to do modular arithmetic, an XOR-like task not expressible by linear models in \mathbbR . We also demonstrate momentum and Adam variants, linear regression, and classification on binary-encoded hierarchies (Quillian semantic networks), addressing open problems posed by Martins (2025). Library at this https URL
[LG-60] Learning Defensive Policies against Diverse Inference Attacks for Smart Meter Privacy
链接: https://arxiv.org/abs/2609.25484
作者: Ruichang Zhang,Mustafa A. Mustafa
类目: Machine Learning (cs.LG)
*备注:
Abstract:Smart meter (SM) data provides fine-grained visibility into household energy consumption, but also exposes users to privacy risks. Inference attacks, known as non-intrusive load monitoring (NILM), can perform appliance-level inference from aggregate signals and recover sensitive behavioral patterns. In practice, attacker models are unknown and heterogeneous, making robust defense challenging. We formulate SM privacy protection as a black-box inference defense problem, aiming to reduce the recoverability of appliance-level information while generalizing across diverse and unseen attackers. We propose a proxy-guided hierarchical reinforcement learning framework that learns battery-based load-shaping policies to inject realistic but misleading appliance-level signatures into the aggregate signal, thereby disrupting the structured patterns exploited by NILM. A self-supervised aggregate-structure privacy probe provides a reconstruction-error-based surrogate reward for disrupting recoverable load structure, while a signature library makes the perturbations appliance-relevant and physically realizable through battery control. We provide theoretical rationale showing that proxy-guided optimization improves inference robustness under attacker diversity. Experiments on real-world datasets UK-DALE and REDD demonstrate strong cross-model and cross-appliance generalization. Across six unseen NILM attackers, covering four appliances on UK-DALE and five on REDD, our proposed defense increases average appliance-level RMSE by 107% and 166%, respectively, while reducing F1 score by 79% and 80%.
[LG-61] WeightBridge: An Efficient Weight Transfer Library for Reinforcement Learning
链接: https://arxiv.org/abs/2609.25442
作者: Xuanlin Jiang,Samuel Hsia,Michael Kuchnik,Zachary DeVito,Minlan Yu,Carole-Jean Wu
类目: Distributed, Parallel, and Cluster Computing (cs.DC); Machine Learning (cs.LG); Networking and Internet Architecture (cs.NI)
*备注: 25 pages, 16 figures
Abstract:Weight transfer - the propagation of updated parameters from trainers to rollout generators - is becoming an important performance bottleneck in reinforcement learning (RL) systems for LLMs. The central challenge is supporting the diverse trainer and rollout layouts and synchronization requirements of modern RL workloads without sacrificing efficiency. Existing solutions are efficient under some configurations but perform poorly or lack support under others. We present WeightBridge, a flexible, efficient weight-transfer library designed to deliver high performance across diverse RL configurations. WeightBridge first automatically extracts the correspondence between trainer and rollout weight layouts, then plans and executes redundancy-free and load-balanced weight transfer. It exposes a small, general API while coordinating workers across diverse synchronization modes. Across configurations spanning different models, parallelization layouts, and synchronization modes, WeightBridge reduces average GPU stall time by up to 42 \times over the state-of-the-art open-source RL framework and achieves high performance in all settings. A coding agent was able to integrate WeightBridge into two different RL frameworks without manual guidance, demonstrating the generality and ease of use of its APIs.
[LG-62] PermuFormer: Multi-Task Pretraining for Permutation Representation in Algebraic Combinatorics
链接: https://arxiv.org/abs/2609.25438
作者: Henry Kvinge
类目: Machine Learning (cs.LG); Combinatorics (math.CO)
*备注: 33 pages. Comments welcome
Abstract:Diverse pretraining has been shown to be an effective method for learning reusable, domain-aware representations that provide a starting point for fine-tuning on downstream tasks. While much of the excitement in AI for math has been concentrated in the use of frontier reasoning models to solve well-specified problems through the medium of language, narrow, specialized models remain an important component of the AI for math ecosystem. In contrast to large language models, specialized models are usually trained directly on the mathematical objects themselves (e.g., graphs, sequences of numbers) rather than the textual descriptions that characterize these objects. However, the common practice of training specialists from scratch may limit their ability to develop domain-aware representations that capture the multifaceted nature of mathematics. In this paper, we describe an approach to pretraining for permutation-focused tasks in algebraic combinatorics. We introduce PermuFormer, an autoregressive transformer trained on a 2.8 billion token multi-task, multi-encoding corpus. We show that PermuFormer is an effective starting point for fine-tuning on basic tasks unseen during pretraining and more complex research-level tasks, frequently outperforming the same architecture trained from scratch, baseline MLPs, and a fine-tuned generic language model of comparable size. We also analyze some of the internal mechanisms by which PermuFormer learns to solve training tasks. For example, we show that while some tasks can be linearly decoded directly from the internal representation of the prompt, other tasks require multiple rounds of generation before the answer can be decoded.
[LG-63] Extending FunctionGemma for Practical On-Device Mobile Function Calling EMNLP2026
链接: https://arxiv.org/abs/2609.25373
作者: Ali Rezagholizadeh,Soheila Samiee
类目: Machine Learning (cs.LG); Software Engineering (cs.SE)
*备注: Accepted at EMNLP 2026. 16 pages, 1 figure, 8 tables. Includes main paper, references, and appendices. Code, datasets, models, and demo available at GitHub and Hugging Face
Abstract:On-device assistants require function-calling models that map natural language to local system actions, but existing resources emphasize web APIs or narrow mobile-action catalogs. We extend FunctionGemma 270M-it to practical Android workflows by introducing MOBILEACTIONSEXTENDED, a synthetic, schema-validated dataset of ~9,500 conversations covering fifteen device-control categories, including messaging, phone calls, camera/screenshot, brightness control, device-status queries, flashlight control, and application management. We fine-tune the 270M model with TRL supervised fine-tuning under completion-only loss, producing an extended specialist and a combined model trained jointly with Google’s MOBILEACTIONSGOOGLE. On MOBILEACTIONSEXTENDED, end-to-end accuracy improves from 29.3% for the base model and 17.2% for Google’s Mobile-Actions variant to 76.5%. The combined model retains 76.5% on MOBILEACTIONSEXTENDED and reaches 82.3% on MOBILEACTIONSGOOGLE, down from the 90.3% of Google’s Mobile-Actions specialist, representing an 8.0-percentage-point trade-off in return for doubling category coverage. We release the dataset, fine-tuned models, reproducible training/evaluation pipeline, and an Android demo, highlighting compact local function calling as a practical path towards low-latency and privacy-preserving mobile assistants.
[LG-64] SSP-Bench: A Hybrid Data Generation Framework for Safety Security and Privacy Evaluation
链接: https://arxiv.org/abs/2609.25352
作者: Fatih Deniz,Yazan Boshmaf,Issa Khalil
类目: Cryptography and Security (cs.CR); Machine Learning (cs.LG); Software Engineering (cs.SE)
*备注:
Abstract:Evaluation of large language models (LLMs) for safety, security, and privacy (SSP) relies heavily on static benchmarks, which suffer from score saturation, data contamination, and aggregation artifacts, and fail to capture sensitivity to linguistic variation. As a result, models that perform well on fixed test sets often fail under semantically equivalent rephrasings. We introduce SSP-Bench, a dynamic benchmarking framework that generates evaluation instances on demand while preserving domain consistency. The framework ensures label validity through externally grounded sources, enforces scope via service-specific validation, and calibrates difficulty using a multi-model steering panel. Benchmark construction is formulated as a multi-objective optimization problem over difficulty, separability, novelty, and diversity. Across 24 models and four SSP services, SSP-Bench reveals systematic failures of static evaluation, including near-zero correlation in safety rankings due to construct mixing, strong safety–over-refusal coupling, and hidden within-family regressions. These results show that static benchmarks can misrepresent model behavior, motivating dynamic, deployment-relevant evaluation.
[LG-65] Concept Drift from a Causal Perspective
链接: https://arxiv.org/abs/2609.25340
作者: Eduardo V. L. Barboza,Jean Paul Barddal,Robert Sabourin,Rafael M. O. Cruz
类目: Machine Learning (cs.LG); Methodology (stat.ME)
*备注:
Abstract:Concept drift is a common phenomenon in real-world data streams, in which changes in the data-generating distribution can degrade predictive model performance. Most existing definitions characterize drift as changes in the joint distribution P(\mathbfx, y) , without distinguishing which component of the data-generating process has changed. In this work, we introduce a causal perspective on concept drift based on Structural Causal Models (SCMs). We propose a taxonomy that categorizes drift events by their causal origin, including changes in exogenous variables, endogenous mechanisms, confounders, and target-generating processes. Building on this framework, we develop an SCM-based data stream generator that simulates controlled mechanism-level drift events. Our experiments empirically characterize the distributional effects of each drift type and show that drifts with different causal origins induce distinct patterns of distribution shift and predictive behavior. Furthermore, by integrating causal discovery methods, we use our framework to construct data streams grounded in real-world dependency structures, enabling more realistic and informative evaluation scenarios. We also demonstrate that leveraging the generated data can improve downstream performance. These results highlight the importance of accounting for causal structure when studying and evaluating adaptive learning methods, and establish a foundation for causally-aware evaluation in non-stationary environments.
[LG-66] GINIO: A Geometric SO(3)-Equivariant Interface for Neural Inertial Odometry
链接: https://arxiv.org/abs/2609.25338
作者: Chankyo Kim,Minghan Zhu,Tzu-Yuan Lin,Avantika Rattan,Maani Ghaffari
类目: Robotics (cs.RO); Machine Learning (cs.LG)
*备注: Accepted at the 10th Conference on Robot Learning (CoRL 2026). 26 pages, 14 figures
Abstract:Neural inertial odometry increasingly uses networks as learned measurements inside filtering pipelines. Such measurements should transform consistently under arbitrary IMU mounting conventions: their mean must transform as a vector, and their covariance must transform congruently as a second-order tensor. We present GINIO, a geometric SO(3)-equivariant interface for neural inertial odometry under arbitrary rotations of the IMU measurement frame. Given calibrated IMU windows, our framework predicts a motion measurement and uncertainty obeying these tensorial laws. To support efficient sensor-frame learning, we introduce Last-Frame Alignment (LFA), a deterministic preprocessing step that is provably equivalent to world-frame training for SO(3)-equivariant predictors. The connected estimator tracks sensor-local states such as IMU bias, separating nuisance estimation from the geometric law enforced by the learned measurement. We instantiate the same interface in filter-connected NIO, AirIO-style recurrent aerial prediction, EqNIO-style full-SO(3) canonicalization, and ResNet-style temporal backbones. On TLIO, GINIO achieves 2.018 m ID/SO(3) ATE while EqNIO degrades to 76.389 m, using 11.6x fewer FLOPs. On NanoBench, our AirIO-style instantiation improves ATE from 5.579 m to 1.430 m without external attitude input, and our ResNet-style instantiation reaches 0.581 m ATE versus 0.645 m for ResNet1D. On Fetch, GINIO empirically reduces unseen physical-remount ATE from 8.15 m to 0.50 m without retraining, demonstrating robustness beyond the exact coordinate-frame guarantee. For uncertainty, spectral covariance reduces covariance-equivariance error by over three orders of magnitude compared with a diagonal head.
[LG-67] Spatiotemporal Kronecker Covariance Neural Networks
链接: https://arxiv.org/abs/2609.25326
作者: Andrea Cavallo,Athanasios Georgoutsos,Elvin Isufi
类目: Machine Learning (cs.LG)
*备注:
Abstract:Multivariate time series contain complex patterns that span across both space and time. While covariance-based statistical tools like spatiotemporal Principal Component Analysis (ST-PCA) help identify these patterns, they are limited to linear operations and prone to estimation errors with limited data. Recent covariance-based spatiotemporal neural networks offer more stable, non-linear alternatives, but they ignore correlations across different time steps. To solve this, we introduce the Kronecker coVariance Neural Network (KVNN), a temporal graph neural network that represents the spatiotemporal covariance matrix via a sum of Kronecker products where spatial and temporal dependencies are decoupled. By implementing filtering operations on spatial and temporal components, KVNNs achieve expressive processing capabilities, admit a rigorous spectral analysis, and are provably stable to finite-sample estimation errors, ultimately addressing all of ST-PCA’s limitations. We show on five real-world datasets that KVNNs achieve strong forecasting performance, often requiring significantly fewer trainable parameters than competitive methods, and are consistent under estimation noise.
[LG-68] opological Signal Processing With Unoriented Operators
链接: https://arxiv.org/abs/2609.25310
作者: Andrea Cavallo,Varun Sarathchandran,Geert Leus,Elvin Isufi
类目: Machine Learning (cs.LG)
*备注:
Abstract:Topological signal processing (TSP) processes signals on simplicial complexes with oriented boundary operators, which is the natural choice for flow signals or when the topological invariants play a role for the task at hand. However, many higher-order signals carry no orientation, and applying oriented operators to them is not well-defined since it introduces an arbitrary choice of simplex orientation. We study an unoriented TSP (UTSP) framework that replaces oriented boundaries with unoriented incidence matrices. First, we show that unoriented incidence and Laplacian matrices between arbitrary simplicial levels admit graph-like spectral properties. Second, since dropping orientation removes the Hodge decomposition, we introduce an unoriented counterpart, termed interaction-order decomposition, which quantifies how much of a higher-order signal is explained by aggregating lower-order signals. Third, we use this decomposition to derive regularizers for signal reconstruction that penalize each interaction order separately. Experiments on real-world data show that the order-aware regularizers outperform oriented baselines, with the largest gains when the signal energy is unevenly distributed across orders.
[LG-69] Correcting Within-Group Self-Selection Bias in Prioritized Replay
链接: https://arxiv.org/abs/2609.25297
作者: Oscar Miró López-Feliu,Herke van Hoof
类目: Machine Learning (cs.LG)
*备注: Accepted at the 19th European Workshop on Reinforcement Learning (EWRL 2026)
Abstract:Prioritized experience replay (PER) improves sample efficiency by replaying high-priority transitions, usually according to absolute temporal-difference error. In stochastic environments, PER can distort the distribution of realized outcomes replayed from transitions with the same state-action pair. We call this within-group self-selection. We quantify the resulting changes in within-group outcome frequencies and mean Bellman targets. We decompose PER into between-group allocation and conditional sibling selection, and derive fixed-buffer corrections that preserve current group-level priority mass: SAMPLE selects a group through PER and trains on a uniformly sampled sibling; AVG averages sibling Bellman targets; and MODEL samples from an empirical full-outcome model. In exact state-action environments with rare high-magnitude outcomes, sibling-aware replay improves learning efficiency over PER, although matched parameter sweeps show that tuning can narrow some gaps. In MinAtar, approximate VQ-VAE groups with SAMPLE mitigate degradation under mean-preserving reward tails in four of five games. Sibling-aware replay thus retains the focus on high-priority state-action regions while recovering their empirical outcome frequencies.
[LG-70] Learning Neural Feedback Linearization for Data-driven Systems via Augmented Lagrangian
链接: https://arxiv.org/abs/2609.25163
作者: Lakshmi Priya P. K.,Andreas Schwung
类目: Machine Learning (cs.LG)
*备注:
Abstract:The paper proposes a novel data-driven framework for designing and training a feedback linearizing controller by explicitly incorporating relative degree based conditions into the learning process. This enables the conventional feedback controller components to be replaced by neural Lie derivatives, thereby facilitating a fully data-driven feedback linearization framework. Furthermore, practical closed-loop stability is established by deriving sufficient conditions under which bounded identification errors lead to bounded tracking errors. The derived theoretical results are validated through their application to an armature controlled DC motor.
[LG-71] Dual-GNN Multilevel Coarsening for Maximum Independent Set
链接: https://arxiv.org/abs/2609.25149
作者: Tianfeng Chen,Xianyue Li
类目: Machine Learning (cs.LG); Combinatorics (math.CO)
*备注: 12 pages, 5 figures, and 6 tables
Abstract:Solving large-scale instances of the Traveling Salesman Problem (TSP) exactly is computationally expensive. Researchers often employ graph sparsification methods to improve computational efficiency. Traditional sparsification methods typically rely on fixed heuristics and fail to fully exploit instance-specific structural information. In this paper, we propose Graph Edge Sparsification (GES), a learning-based sparsification approach for Euclidean TSP. By incorporating geometric structural information and combinatorial optimization technology, our proposed method adaptively generates a sparsification graph for different instances, significantly reducing the graph size and accelerating the solving process. Experimental results demonstrate that our sparsification method can prune up to 95% of edges on the MATILDA dataset, while keeping the solution gap within 1% of the optimal value. Moreover, our approach exhibits strong generalization capability on the TSPLIB this http URL some large-scale instances, the pruning rate exceeds 99%, while the optimality gap remains below 1%.
[LG-72] Entropy Can Flow or It Can Guide. Be Entropy. LEDFlow: Introducing Entropy-guided Generation Order into Uniform Discrete Flow
链接: https://arxiv.org/abs/2609.25131
作者: Tung Sum Thomas Kwok,Yidong Ouyang,Yingjia Wan,Ying Nian Wu,Zhijiang Guo,Oscar Leong
类目: Machine Learning (cs.LG)
*备注:
Abstract:Uniform discrete flow permits repeated updates at every generation position. While continued revision supports correction of wrong tokens, it also exposes correct intermediate predictions to later errors. An experiment on Sudoku puzzles shows that 9.4% of generated cells are correct at an intermediate step but incorrect in the final output. We introduce generation order into uniform discrete flow through selective absorption, which fixes chosen predictions while preserving the uniform-flow velocity at active positions. To prevent absorbing incorrect predictions, we propose Low-Entropy Discrete Flow (LEDFlow), a training-free sampler that adaptively orders absorption by local entropy. By decomposing absorption error into joint dependence and conditional prediction terms, we show that selecting the lowest-entropy positions under a fixed absorption budget minimizes an upper bound on the conditional term. We further support the choice of local entropy by showing that the decision-error bound of global lookahead grows with the lookahead window under an imperfect denoiser. Across reasoning benchmarks, LEDFlow attains 0.845 Nikoli Sudoku solve accuracy, with the largest gains on strongly constrained tasks. On text-to-image generation it attains the best overall score, and on multimodal understanding it improves over the native sampler on all six benchmarks, at an inference cost comparable to standard flow sampling.
[LG-73] A Decentralized Partially Observable Team Decision Methodology with Delayed Information Sharing
链接: https://arxiv.org/abs/2609.26783
作者: Xiaoxing Ren,Thomas Parisini,Andreas A. Malikopoulos
类目: Optimization and Control (math.OC); Machine Learning (cs.LG); Systems and Control (eess.SY); Machine Learning (stat.ML)
*备注: 15 pages, 2 figures
Abstract:We study decentralized partially observable team decision problems with low-rank latent dynamics and unknown system models. The proposed framework combines team-theoretic equivalence with low-rank model representations to address cooperative decision-making in partially observable Markov decision processes without prior knowledge of the transition model. Each team member makes decisions based on local private information and delayed common information shared across the team. Using only this available information, each member learns an approximate low-rank Markov decision process and applies least-squares value iteration to compute its policy. This yields a fully decentralized learning and planning algorithm that requires neither a centralized coordinator nor centralized training. We show that the resulting member-side solutions approximate the centralized team solution: despite partial observability, unknown dynamics, and delayed common information, each member recovers the corresponding component of an approximate team-optimal policy. We further establish finite-sample performance guarantees and derive a corresponding sample-complexity bound for the proposed algorithm.
[LG-74] Automatic depth-based local center clustering via β-integrated local depth and adaptive grouping
链接: https://arxiv.org/abs/2609.26748
作者: Siyi Wang,Alexandre Leblanc,Paul D. McNicholas
类目: Methodology (stat.ME); Machine Learning (cs.LG); Machine Learning (stat.ML)
*备注:
Abstract:Clustering is an unsupervised learning technique that partitions unlabeled data into groups. Most existing methods require user-specified parameters, such as the number of clusters or neighborhood size. Conversely, we propose automatic depth-based local center clustering (A-DLCC), a fully data-driven method that eliminates numerical parameter tuning. A-DLCC uses the \beta -integrated local depth to identify stable exemplars, points consistently central across multiple locality levels, termed local centers, which are ranked by their representativeness. Each local center induces a group of similar points, with group-level similarity measured by a proposed nonparametric metric called group-level local similarity. To guide merging, we incorporate the bottleneck path idea from graph theory, which forms the basis of our adaptive merging criterion. Based on this criterion, we design a single agglomeration rule in which a group is either absorbed by a neighbor it reaches better than itself or bonded to a neighbor that both sides find more reachable than their own background, every merge being additionally required to be carried by a contact stronger than a configuration-model null expects. The rule automatically estimates the number of clusters and decides when to stop merging. Experiments on synthetic and real data show that A-DLCC produces interpretable clustering results without parameter tuning.
[LG-75] Optimal Sequential Annotations for Off-Policy Evaluation
链接: https://arxiv.org/abs/2609.26707
作者: Woojin Chae,Ezinne Nwankwo,Haitong Qin,Angela Zhou
类目: Methodology (stat.ME); Machine Learning (cs.LG); Machine Learning (stat.ML)
*备注:
Abstract:Offline reinforcement learning and off-policy evaluation evaluates dynamic treatment rules based on retrospectively collected data prior to deployment. In recent AI applications, state and reward information is recorded as complex text or image, which recent AI advancements such as LLM-as-a-judge can label with unknown bias. Expert annotation may be available but at a higher cost. For example, safety classification via cheap but imperfect classifiers vs. expensive expert review. We show how a limited budget for ground-truth data-annotation can be used via doubly-robust OPE with missing rewards, and we optimize variance-optimal annotation probabilities for sequential off-policy evaluation, where the target policy value is estimated from annotated data. We characterize the optimal annotation probabilities for sequential forward-monotone annotation protocols, and provide a feasible batch-adaptive implementation. Our work is motivated by a collaboration with a homelessness services nonprofit that writes casenotes for individuals over time. Our method can be used to unlock trustworthy inference from casenote data and answer new inferential questions such as: how does expanding outreach effort over time affect progress towards a housing application and improvement in housing placement? In simulations and on two real datasets - casenotes from the nonprofit and human-preference votes from LMArena - we see reductions in RMSE of 34-65% for housing placement and 17-68% for progress towards a housing application at budgets of 40% of full annotation and above, and by 55-62% at every budget on LMArena.
[LG-76] When are bosonic Gaussian states classical to learn?
链接: https://arxiv.org/abs/2609.26705
作者: Senrui Chen,Antonio Anna Mele,Francesco Anna Mele,John Preskill
类目: Quantum Physics (quant-ph); Information Theory (cs.IT); Machine Learning (cs.LG); Mathematical Physics (math-ph)
*备注: comments welcome
Abstract:A fundamental question in physics is: When does classical behavior emerge from quantum systems? Bosonic Gaussian states provide a natural setting to explore this quantum-classical boundary, as they capture both the classical field behavior and the intrinsic quantum nature of light. Here, we address this problem from a learning-theoretic perspective by asking: When are bosonic Gaussian states classical to learn? That is, under what conditions (if any) can an n-mode bosonic Gaussian state be learned with as few samples, and with operations as simple, as are needed to learn a classical 2n-variate Gaussian distribution? We establish a smooth crossover in learnability governed by the state’s thermal fluctuations: - Cold Gaussian states are non-classical to learn: When the covariance matrix satisfies \Sigma\le(\frac12+O(\frac1n))I , i.e. close to the vacuum covariance, tomography under single-copy (i.e., non-entangled) measurements fundamentally requires \Omega(n^3) copies, strictly exceeding the sample complexity \Theta(n^2) of learning classical Gaussian distributions. We show that this hardness persists even when few-copy entangled measurements are allowed. - Warm Gaussian states are classical to learn: When thermal fluctuations exceed the vacuum noise, parameterized by \Sigma\ge(\frac12+\nu)I for any parameter \nu0 , we prove that single-copy tomography requires N=\Theta\left(n^2\min(n,1+\nu^-1)\right) copies. This bound is tight and is achieved by simple, non-adaptive, unentangled heterodyne measurements. Crucially, for \nu=\Omega(1) , the sample complexity drops to \Theta(n^2) , matching the classical case. Our results tightly characterize a quantum-to-classical crossover in the learnability of bosonic Gaussian states, reveal a novel connection between fundamental physics and statistical learning theory, and have implications for real-world sensing experiments. Comments: comments welcome Subjects: Quantum Physics (quant-ph); Information Theory (cs.IT); Machine Learning (cs.LG); Mathematical Physics (math-ph) Cite as: arXiv:2609.26705 [quant-ph] (or arXiv:2609.26705v1 [quant-ph] for this version) https://doi.org/10.48550/arXiv.2609.26705 Focus to learn more arXiv-issued DOI via DataCite (pending registration) Submission history From: Senrui Chen [view email] [v1] Tue, 22 Sep 2026 16:58:25 UTC (191 KB)
[LG-77] PROSWIN: Probabilistic Solar Wind Speed Forecasting Using Deep Distributional Regression From Solar Images
链接: https://arxiv.org/abs/2609.26683
作者: Daniel Collin,Yuri Shprits,Luca Chiarabini,Stefan J. Hofmeister,Nadja Klein,Guillermo Gallego
类目: olar and Stellar Astrophysics (astro-ph.SR); Machine Learning (cs.LG); Data Analysis, Statistics and Probability (physics.data-an); Space Physics (physics.space-ph)
*备注:
Abstract:Accurately predicting fast solar wind conditions is challenging, as uncertainties are large and unquantified by traditional single-value prediction models. In particular, the risks of high-speed solar wind streams (HSSs), which can cause damage to technological infrastructure, cannot be reliably assessed without probabilistic forecasts. We present PROSWIN, a probabilistic machine learning model that forecasts the hourly solar wind speed (SWS) at Earth with a four-day lead time. The approach combines solar images and magnetograms using a deep neural network coupled to a distributional regression algorithm. Because standard error metrics underweight the relevance of HSS peaks, we further introduce the prediction score, a model-selection metric that jointly rewards timeline and HSS peak accuracy. On 14 years of data, our forecast achieves very well-calibrated uncertainties (1% average deviation). Using the continuous ranked probability score (CRPS), a metric that assesses distributional accuracy, we obtain a timeline CRPS of 41.0 km/s, an HSS peak CRPS of 45.3 km/s, and a prediction score of 42.3 km/s. We find that the 171 Å channel is an important complement to the typically used 193 Å and 211 Å channels and that the prediction score for model selection improves the applicability of the model. Compared to selected models from the literature, ours is the only one that is accurate for both timeline and HSS peak values, rather than trading one off against the other. These results support the advantages of probabilistic over single-value solar wind models. The introduced methods are also transferable to other forecasting problems.
[LG-78] Statistical Rates for Entropic Optimal Transport in the Discrete to SubGaussian Regime
链接: https://arxiv.org/abs/2609.26647
作者: Tomas Gonzalez,Gonzalo Mena
类目: atistics Theory (math.ST); Machine Learning (cs.LG)
*备注:
Abstract:We study statistical rates in entropic optimal transport in the semi-discrete regime where one measure has finite support and the other is subGaussian. Our main result establishes parametric convergence rates for the empirical dual potentials to their population counterparts, with no dimension dependence in the leading term. Our result relies on tailored strong concavity analysis of the semi-dual objective, coupled with specialized bounds for the semi-discrete potentials. As a consequence, we obtain fast rates for downstream quantities derived from the optimal coupling. Chiefly, the empirical barycentric projection achieves a squared-error rate n^-1 , matching the fully compact case and improving over the less favorable n^-1/2 rate known for fully subGaussian settings. Altogether, these results may indicate a lower complexity adaptation phenomenon whereby the statistical complexity of the barycentric projection is governed by the discrete measure. As an application, we analyze Sinkhorn-EM, an EM-type algorithm in which the E-step is replaced by an entropic optimal transport problem. In a well-specified and balanced two-component Gaussian mixture model, we prove \sqrtn -consistency of the empirical iterates to their population counterparts for any fixed number of iterations, matching classical EM rates up to a \sqrt\log n factor. Simulations support the theory.
[LG-79] On Basis Function Selection for Sparse Gaussian Process Regression
链接: https://arxiv.org/abs/2609.26624
作者: Marnix Van Soom,Ivan De Boi
类目: Machine Learning (stat.ML); Machine Learning (cs.LG)
*备注: 19 pages, 8 figures
Abstract:Sparse Gaussian processes achieve O(N) inference by replacing the kernel with an appropriate expansion in a fixed basis \phi_j\ on the input space. Given a compute budget M \ll N , practitioners conventionally truncate the basis to its first M entries. Nothing in the formalism, however, prevents one from selecting only those M basis functions that matter for the data at hand. This would avoid spending budget on basis functions where there is no signal, but it requires a criterion for ranking the candidates. We propose three such criteria derived from an information-theoretic view of the basis-function selection problem. Each criterion matches a different state of knowledge at selection time: a no-data state, a no-prior state, and an in-between state. We then study the performance of truncation versus selection strategies on six UCI regression benchmarks across three basis families: Hilbert-space Gaussian processes (HSGP), variational Fourier features (VFF), and variational inducing spherical harmonics (VISH). We observe that the no-data criterion is a safe default, matching or improving on truncation for HSGP, VFF and VISH, with substantial gains for VISH and improvements over a recently developed selection heuristic for that basis family. The data-aware no-prior and in-between criteria provide substantial gains over truncation specifically for HSGP, which is the most broadly used of the three families in practice.
[LG-80] Unlocking Cross-Scenario Physical Layer Security: A Mixture-of-Experts Framework with Generative Diffusion Models
链接: https://arxiv.org/abs/2609.26598
作者: Xiao Tang,Tong Hui,Chao Shen,Yichen Wang,Qinghe Du,Li Sun,Zhu Han
类目: ignal Processing (eess.SP); Information Theory (cs.IT); Machine Learning (cs.LG)
*备注: Accepted @ IEEE TIFS
Abstract:The future 6G networks are expected to incorporate a proliferation of wireless services in diverse environments, which presents a significant challenge for information security. Conventionally optimization always requires recalculation and learning strategy often suffers poor generalization, which are thus incapable for the security provisioning with wide scenario coverage. In this paper, we propose an adaptive and robust learning framework that leverages a mixture-of-experts (MoE) architecture to achieve cross-scenario physical layer security guarantee. Specifically, we first select a few representative scenarios and establish the scenario-specific generative diffusion model (GDM)-based experts for secure transmission beamforming with artificial noise. The diffusion nature of experts learns the overall probability distribution of security strategy solution landscape and the Transformer-based denoising process enhances the ability to generalize across varying network configurations. Then, a lightweight gating network is constructed to identify the scenarios by engineering the channel features and select the most relevant experts. Finally, an attention-based combiner is introduced to synthesize the security proposals from the top-rated experts to produce a high-fidelity security strategy to cover the unseen scenarios. Simulation results demonstrate that the proposed GDM-based MoE framework can accurately recognize the scenarios and properly select the experts, maintaining near-optimal secrecy rates across a continuum of wireless scenarios and outperforming traditional single-model paradigms.
[LG-81] Polyak-Type Extrag radient Methods for Monotone Root-Finding Problems
链接: https://arxiv.org/abs/2609.26581
作者: TaeHo Yoon,Sayantan Choudhury,Ezra Greenberg,Nicolas Loizou
类目: Optimization and Control (math.OC); Machine Learning (cs.LG)
*备注:
Abstract:We study Polyak-type step-size selection for extragradient methods for solving deterministic and stochastic monotone root-finding problems. We show that the known projection-type correction for deterministic extragradient arises from minimizing an upper bound on the distance to a solution, paralleling the classical Polyak step-size construction. Using this viewpoint, we provide a unified deterministic analysis of the Polyak-type Extragradient Method (PolyakEG), based on a local critical condition controlling the variation of operator F along the extrapolation direction. This analysis does not require global Lipschitz continuity, and covers sublinear convergence under broader conditions such as Hölder continuity or (L_0, L_1) -Lipschitzness and linear convergence under additional strong monotonicity, all through a single framework. We then study the stochastic extensions of this approach. We first prove convergence of a direct stochastic variant, PolyakSEG, when all stochastic component operators share a common solution. We also show that, without this condition, PolyakSEG with nonvanishing step-sizes may fail to converge to a zero of the mean operator. To address this limitation, we propose DecPolyakSEG, which combines decreasing step-sizes with Polyak-type updates, and establish a sublinear residual convergence result without requiring a common solution across the component operators. These results parallel recent developments in stochastic Polyak step-sizes from the convex minimization literature and establish an analogous research avenue in the broader root-finding regime.
[LG-82] Deep Generative Crystal Structure Prediction: A Benchmark Study and a Controlled Test of Prototype Dependence
链接: https://arxiv.org/abs/2609.26502
作者: Lai Wei,Rongzhi Dong,Ying Feng,Madeline Miklos,Jianjun Hu
类目: Materials Science (cond-mat.mtrl-sci); Machine Learning (cs.LG)
*备注: 18 pages
Abstract:Deep generative models are widely reported to enable de novo crystal structure prediction (CSP), but their capability has not been measured consistently against template-based methods. We evaluate 12 representative generative CSP models, spanning latent-variable, diffusion, flow-matching, autoregressive, and manifold random-walk architectures, against TCSP 2.0 on 180 test structures and a leakage-controlled subset of 46. All methods use identical structure-matching, symmetry, and consensus criteria. Template retrieval is the strongest single method, reaching 68.3% top-1 success; symmetry-aware EquiCSP (66.4%) and Uni-3DAR (62.9%) form the next tier. However, comparison with TCSP 2.0 shows that most structures correctly predicted by generative models are also correctly predicted by template substitution. Thus, the set of structures uniquely reachable by generation is small, limiting its practical advantage for discovering structures outside existing prototype libraries. To test the source of this performance, we removed entire stoichiometric prototype families from the training set and retrained the strongest generative model. Accuracy declined by 50-78% across four families, establishing that performance is substantially prototype-dependent. A small minority of structures survived removal of their prototype family, demonstrating a real but limited retrieval-independent predictive capacity. Present generative CSP models therefore function largely as implicit, softer-edged prototype libraries rather than genuinely de novo predictors. Enlarging this residual capacity, rather than aggregate match rate alone, is the central open problem.
[LG-83] A Practical Guide on Graphical Model Validation
链接: https://arxiv.org/abs/2609.26445
作者: Mario V. Wüthrich
类目: Machine Learning (stat.ML); Machine Learning (cs.LG); Risk Management (q-fin.RM)
*备注:
Abstract:This manuscript formalizes the most popular model validation tools used in general insurance actuarial modeling. These include graphical tools like calibration plots, actual-vs-expected plots, lift charts, Murphy diagrams, as well as classical statistical tools such as Bregman losses, deviance losses, elementary losses, Murphy’s decomposition and Gini scores. Particular emphasis is placed on whether calibration and discrimination are studied under a policy-weighted or an exposure-weighted population measure. This distinction is crucial in ensuring that premium schemes are calibrated on the correct scale.
[LG-84] SuperPCA: subspace analysis and an efficient algorithm for high-dimensional PCA
链接: https://arxiv.org/abs/2609.26406
作者: Irina-Beatrice Haas,Maike Meier,Yuji Nakatsukasa,Taejun Park
类目: Machine Learning (stat.ML); Machine Learning (cs.LG); Numerical Analysis (math.NA); Computation (stat.CO)
*备注: 22 pages, 8 figures
Abstract:Principal component analysis (PCA) is a fundamental tool to reduce the dimensionality of the data in many applications. PCA finds a few signal directions that contain most of the variability of the data by computing the eigenvectors of the sample covariance matrix. In this work, we focus on the spiked covariance model, in which the data vectors are defined by a few orthogonal signals plus an isotropic Gaussian noise, and our goal is to estimate one or more of the leading signals. Our main theoretical finding is that the subspace spanned by several leading eigenvectors of the sample covariance matrix contains significant information about the desired signals long before the individual eigenvectors converge to the population principal components. To prove this, we derive a posteriori bounds for the angle between the subspace spanned by the desired population signals and the subspace obtained from the sample using perturbation theory for singular vectors. This leads to a new algorithm, SuperPCA (SUbsPace subsamplER PCA), which capitalizes on an approximate eigenspace of the sample covariance matrix to find the leading signals far more efficiently and accurately than classical PCA in the high-dimensional, multi-signal setting. SuperPCA exploits only a small number of subsampled coordinates of the data, which can lead to tremendous savings in data acquisition cost, especially when the signals are approximately sparse. For the same number of measurements, SuperPCA can offer a factor 10 improvement in accuracy compared to the classical PCA method.
[LG-85] Error Bounds for Statistical Estimators in BTL Model with Parametric Multivariate Utility Functions
链接: https://arxiv.org/abs/2609.26326
作者: Yicheng Li,Huifu Xu
类目: Machine Learning (stat.ML); Machine Learning (cs.LG)
*备注:
Abstract:We study preference elicitation under the Bradley-Terry-Luce (BTL) model where the true partworth vector is unknown and has to be estimated as a parameter with elicited preference information. The set of selected pairwise queries is non-uniform, deterministic, and arbitrary over a collection of alternatives, provided that it satisfies a joint identifiability condition. We focus on understanding when the canonical maximum likelihood estimator (MLE) is finite and admits sharp error bounds without explicit compactness constraints on the feasible set or external regularizers. To this end, we derive minimax lower bounds under the standard bounded dynamic range condition, and find that the same Fisher-information geometry in the classic Cramér-Rao lower bounds underpins the finite-sample difficulty of the estimation problem. By combining a non-asymptotic expansion of the likelihood score equation with a fixed-point localization argument, we identify a design-dependent sample size threshold above which the unconstrained canonical MLE exists and is unique with high probability. The same expansion yields a decomposition of the estimation error into a linear stochastic term, an explicit second-order bias, and a higher-order remainder. A refined analysis gives sufficient sample size conditions under which the canonical MLE attains the minimax rates up to logarithmic and constant factors. These results provide a unified non-asymptotic theory for parametric utility elicitation and reveal when the inference is determined by response data alone rather than by external regularization. Preliminary numerical results are consistent with the theoretical findings.
[LG-86] arget alignment dilution and forecast selection when cross-sectional forecasts share a common target
链接: https://arxiv.org/abs/2609.26303
作者: Masoud Soleimani
类目: Econometrics (econ.EM); Machine Learning (cs.LG); Statistical Finance (q-fin.ST); Methodology (stat.ME)
*备注: 35 pages, 5 figures, 13 tables
Abstract:Forecasters often score the same units per date against one standardized realized outcome. We show that every standardized forecast splits exactly into a component aligned with this common target and a component uncorrelated with it. Three consequences follow: forecast-error correlation largely mirrors forecast correlation and is therefore a poor measure of diversity; an equally weighted combination beats a no-information forecast only when average alignment is large relative to the combination’s dispersion; and the gain from adding a forecaster separates into genuine improvement and mere dilution, which equal-weight admission can mistakenly reward. We develop a cautious selection rule, study it in simulations, and apply it to language-model forecasts of US equity rankings and mechanical signals ranking exchange-traded funds. Selection removes most dilution losses, but no combination beats the no-information forecast.
[LG-87] Learning to Fluctuate: Statistical Foundations for Causal Tabular Pretraining
链接: https://arxiv.org/abs/2609.26290
作者: Zhiheng Zhang
类目: Machine Learning (stat.ML); Machine Learning (cs.LG)
*备注:
Abstract:Causal tabular foundation models amortize effect estimation across synthetic mechanisms, but latent-effect supervision rewards posterior shrinkage instead of directly encoding the repeated-sample response needed in a fixed deployment population. We introduce fluctuation-supervised pretraining (FSP): each synthetic table is labeled by its average treatment effect plus its efficient influence-function fluctuation, while deployment remains a single frozen forward pass. Along the path T_\lambda,P=\theta§+\lambda P_n\psi_P , we prove an endpoint transition: every fixed \lambda1 retains label ambiguity of order (1-\lambda)^2/n , whereas full fluctuation makes the Gaussian label observable and reduces optimal finite-stratum causal label-prediction risk to order n^-2 . One finite-pretraining bound combines label, network, episode-sampling, and optimization errors; its resulting sampling defect controls fixed-mechanism bias, mean squared error, variance, Gaussian approximation, and, with variance-head accuracy, studentized coverage. Complementary lower bounds separate the local n^-1 ATE risk that deployment observations cannot erase from the \log N/M excess risk of a generic finite-dictionary episode-learning problem. Experiments trace the learned sampling response. With a raw-row/column backbone, FSP reduces large-effect-shift RMSE by 69.8% relative to latent supervision and by 39.5% relative to a released CausalPFN checkpoint on matched tables. Continuous-covariate experiments, known-effect semisynthesis and two randomized-study evaluations separate sampling-law fidelity from point-risk shrinkage and expose weak-overlap errors in both learned heads.
[LG-88] Sample-Smooth Spaces: A Convenient Category for Differentiable Probabilistic Programming
链接: https://arxiv.org/abs/2609.26270
作者: Patrick Forré
类目: Category Theory (math.CT); Machine Learning (cs.LG); Logic in Computer Science (cs.LO); Programming Languages (cs.PL); Probability (math.PR)
*备注:
Abstract:We introduce the category \mathbfSSS of sample-smooth spaces over a mixed site. The test objects are the products \Omega_n := \mathbbR^n \times \Omega of a Cartesian space with the universal Hilbert cube \Omega carrying all universally measurable sets, and a space is a set with a family of admissible plots \Omega_n \to \mathcalX closed under precomposition. Smoothness and measurability are then not two structures glued along an axiom, but one structure over one site. The site has finite non-empty products, because \Omega absorbs its own square; its Karoubi envelope contains every \mathbbR^n ; and it has mixed morphisms \omega \mapsto (W(\omega),\Phi(\omega)) , which turn measurability of a smooth family from an axiom into a consequence. \mathbfSSS is a concrete quasitopos: complete, cocomplete, cartesian closed and locally cartesian closed, with a classifier for embeddings. Morphisms of Cartesian spaces are exactly the C^\infty maps and manifolds embed full and faithfully, both without Boman’s theorem. Every object has tangent and cotangent spaces, every morphism a differential. The modalities sit in an adjoint string \Pi \dashv \flat \dashv \natural \dashv \sharp \dashv \Lambda , making \mathbfSSS cohesive over quasi-universal spaces. The point is the probability monad. Defining the plots of \mathsfP(\mathcalX) as push-forwards of \mathcalX -plots at every test object, \mathsfP is an unconditional strong commutative affine monad on all of \mathbfSSS – functor, unit, product of kernels, multiplication and the monad laws are each one line of seed splitting – and its Kleisli category, of differentiable simulators, is a Markov category. The reparametrisation trick holds by construction: every Kleisli morphism is plot-wise a sampler, stably under composition. A reflection theorem locates the whole gain in a single plot family. Subjects: Category Theory (math.CT); Machine Learning (cs.LG); Logic in Computer Science (cs.LO); Programming Languages (cs.PL); Probability (math.PR) MSC classes: 18B25, 18C20, 60A05 (Primary), 18C15, 18F20, 18M05, 58A40, 28A05, 68Q55, 68N18, 03B38 (Secondary) Cite as: arXiv:2609.26270 [math.CT] (or arXiv:2609.26270v1 [math.CT] for this version) https://doi.org/10.48550/arXiv.2609.26270 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
[LG-89] Hyperbolic Restricted Boltzmann Machine Neural Quantum State
链接: https://arxiv.org/abs/2609.26032
作者: H. L. Dao
类目: Quantum Physics (quant-ph); Disordered Systems and Neural Networks (cond-mat.dis-nn); Machine Learning (cs.LG)
*备注:
Abstract:We construct the first type of non-Euclidean non-autoregressive neural quantum state (NQS) in the form of the hyperbolic Restricted Boltzmann Machine (HRBM), which is studied in the variational Monte-Carlo (VMC) setting of the Quantum Sherrington-Kirkpatrick (QSK) model whose ground state exhibits volume-law entanglement. Across a 512-fold increase in the Hilbert space dimension corresponding to a system size increase from N=14 to N=24 , HRBM NQS robustly outperforms its Euclidean version, the RBM NQS, in terms of better ground state energy optimization as well as lower Renyi-2 S_2 and von Neumann S_vN absolute entanglement entropy reconstruction errors. More importantly, for all QSK system sizes, HRBM NQS demonstrates a superior expressivity in faithfully reproducing the entire entanglement spectrum of the QSK model from the top eigenvalues down to the tail end across 15 orders of magnitude, while RBM NQS consistently overestimates the sub-dominant modes. This work furnishes a proof-of-concept demonstrating that hyperbolic non-autoregressive NQS ansatzë, thanks to the exponential volume of the hyperbolic geometry underlying their constructions, might be more natural at representing volume-law quantum systems than conventional Euclidean NQS. Furthermore, an interesting byproduct of this work is the polynomial scaling result of RBM-type NQS ansatzë in the QSK volume-law system as the Hilbert space increases exponentially.
[LG-90] Bridge of Ψs: Quantum Circuit Optimization with Schrödinger Bridges
链接: https://arxiv.org/abs/2609.25947
作者: Lino S. Hofstetter,Lia Yeh,Prakash Murali
类目: Quantum Physics (quant-ph); Machine Learning (cs.LG)
*备注: 25 pages, 11 figures, 12 tables
Abstract:Quantum circuit optimization replaces a circuit with an equivalent one of fewer gates and lower depth, reducing execution cost and error rate. We ask whether a generative model can learn this transformation directly from examples, rather than selecting from a fixed rewrite library or rigid algebraic routines. We present Bridge of \Psi 's (BOPS), a generative model based on Schrödinger bridges, using a custom denoiser architecture, that learns a transformation from a source circuit into an equivalent optimized circuit. We train it on data constructed to be hard for existing optimizers, by applying rewrite rules backwards so that each input has a known lower-cost target. On held-out 8 qubits \times 64 depth Clifford+ T circuits, BOPS reduces gate count by 2.46\times and depth by 2.45\times in geometric mean, outperforming all nine baseline optimizers. This constitutes the first generative model bridging quantum circuits and frontier machine learning methods, opening up the quantum compilation stack to learned optimization along multiple axes.
[LG-91] Conditional Tensor Diffusion: Distributional Counterfactual Learning and Inference
链接: https://arxiv.org/abs/2609.25924
作者: Xinbing Kong,Zeyu Li,Junfan Mao,Bin Wu
类目: Machine Learning (stat.ML); Machine Learning (cs.LG); Econometrics (econ.EM)
*备注:
Abstract:Causal inference guides operational and managerial decisions but remains challenging in high-dimensional panel or tensor settings, where decisions may depend on the joint conditional distribution of missing control outcomes. We develop \emphCounterfactual Tucker Diffusion (\CFTDiff), which integrates the treatment mask and latent Tucker structure into conditional diffusion to recover this distribution given observed control outcomes through efficient nonlinear score learning in a low-dimensional core. The masked Tucker score preserves dependence across tensor modes while reducing the dimension of nonlinear score learning from the product of mode dimensions to the much smaller product of Tucker ranks. We establish high-probability error bounds for conditional score estimation that depend on the Tucker ranks, largest mode dimension, and the factor-strength-adjusted number of missing outcomes, and show how these bounds translate into recovery guaranties for the conditional distribution of the missing control outcomes. Across missing rates, simulations show more accurate point recovery than common causal panel and matrix/tensor completion methods; comparisons with nested diffusion specifications further demonstrate the gains from masked conditioning and Tucker dimension reduction. In Norway’s iFlex experiment, \CFTDiff recovers missing outcomes more accurately than competing methods; when applied to causal analysis, its estimated conditional distributions yield counterfactual prediction intervals and target-attainment probabilities, allowing pricing interventions to be evaluated by demand-reduction magnitude and reliability.
[LG-92] Statistical Gains from Looped Estimation under Parameter Budgets
链接: https://arxiv.org/abs/2609.25778
作者: Xinyu Tian,Xiaotong Shen
类目: Machine Learning (stat.ML); Machine Learning (cs.LG); Statistics Theory (math.ST)
*备注: 57 pages, 5 figures
Abstract:Growing memory demands in artificial intelligence motivate learning with fewer trainable parameters. We ask whether a looped estimator, which repeatedly applies one fitted operator with parameters shared across iterations, can improve statistical accuracy under a common parameter budget. Its conventional untied counterpart uses separate parameters at each iteration. For general likelihood models, we establish an upper bound on squared Hellinger risk for looped sieve maximum likelihood and a minimax lower bound over the tuned untied family. These bounds reveal a parameter–iteration–accuracy tradeoff: repeated computation can improve approximation without adding parameters, while increasing computational cost and fitted-class complexity. For targets of known Hölder smoothness, looped residual feedforward networks and a specified post-layer-normalized Transformer attain the minimax polynomial rate up to logarithmic factors with a fixed number of bounded real parameters. At sufficiently large fixed budgets, looped worst-case risk vanishes as sample size grows, whereas optimal worst-case untied risk remains bounded away from zero. Under specified growing-budget conditions, the loop-to-untied risk ratio also tends to zero. Gaussian and Laplace regression, binary response, and energy-based density estimation illustrate the theory.
[LG-93] Graded Representation Theory of Equivariant Neural Networks
链接: https://arxiv.org/abs/2609.25776
作者: Mani Shayestehfar
类目: Representation Theory (math.RT); Machine Learning (cs.LG)
*备注: 33 pages, comments welcome
Abstract:Nonlinear activations can create equivariant interactions between irreducible representations that linear maps cannot. We use the Gaussian degree decomposition to extend ordinary polynomial degree to such nonlinear maps, and prove that for a fixed coordinatewise equivariant layer each degree factors into a polynomial determined by the linear maps and a scalar determined by the activation. This separates three distinct obstructions, coming from symmetry, coordinates, and activation.
[LG-94] Optimal Tradeoffs Between Network Size and Parameter Magnitude in Neural Approximation and Minimax Regression
链接: https://arxiv.org/abs/2609.25710
作者: Baicheng Li,Zuowei Shen,Haizhao Yang,Shijun Zhang
类目: Machine Learning (stat.ML); Machine Learning (cs.LG)
*备注: 71 pages
Abstract:The statistical accuracy of neural networks depends on both their approximation power and the complexity of the class fitted from data. While increasing network size is a natural way to improve approximation, parameter magnitude provides another resource whose role must be quantified in both respects. We establish a sharp width–magnitude tradeoff at fixed depth using one elementary bounded 1 -Lipschitz Dyadic–Triangular Activation. For the unit \beta -Hölder ball on [0,1]^d with 0\beta\leq1 , the optimal L^p approximation error for 0p\infty is of order [N^2\log(eNT)]^-\beta/d when the network width satisfies N\geq2d+3 and the parameter magnitudes are bounded by T\geq1 . Matching lower bounds hold for every fixed globally Hölder activation; its Hölder exponent affects the constants but not the rate. Under bounded design densities and independent centered sub-Gaussian noise, approximate least squares over the full clipped class at depth 23 attains the classical Hölder minimax risk \mathcalO(M^-\frac2\beta2\beta+d) without logarithmic loss whenever N^2\log(eNT)\asymp M^\fracd2\beta+d , where M is the sample size. This yields a continuum of statistically optimal choices, ranging from unit parameter radius to fixed network size. At fixed size, four hidden layers with at most 8d+7 nonzero parameters give a near-optimal radius, while six layers with at most 8d+27 attain the optimal order \log T=\mathcalO(\eta^-d/\beta) at approximation error \eta . The same decoding method also yields fixed-size Transformer approximation.
[LG-95] On the Gradient Heterogeneity Dynamics of Adversarially Robust Federated Regression
链接: https://arxiv.org/abs/2609.25705
作者: Leonardo F. Toso,James Anderson,Nirupam Gupta,Rafael Pinot
类目: Machine Learning (stat.ML); Machine Learning (cs.LG)
*备注:
Abstract:Federated learning (FL) is intrinsically heterogeneous: honest clients may have different data-generating models. On top of that, adversarial clients can make heterogeneity even more pronounced by sharing arbitrary updates. Existing analyses typically control the interaction between statistical heterogeneity and adversarial behavior through gradient-dissimilarity conditions. However, the underlying bound is imposed a priori and may yield conservative guarantees even for least-squares regression. We instead derive the gradient heterogeneity from the statistical model of linear and nonlinear regression with fresh data samples at every round. Our bounds separate heterogeneity among the honest clients’ ground-truth model parameters, finite-sample label noise, and initialization. We then demonstrate that, for any (f,\kappa) -robust aggregator with coefficient \kappa = O(f/n) , where f is the number of adversarial clients and n the total number of clients (with f/n 1/2 ), convergence holds after an explicit sample burn-in.
[LG-96] Generalized Deep Regression for Repeated Measurements
链接: https://arxiv.org/abs/2609.25605
作者: Kexuan Li
类目: Machine Learning (stat.ML); Machine Learning (cs.LG)
*备注:
Abstract:In this paper, we study the estimation of a marginal regression function from independent units with repeated binary, count, or continuous responses using ReLU deep neural networks. In the model, we assume that the dependence is generated by an unobserved random mean function within each unit. We then fit a neural network with a convex generalized regression loss. We show an oracle inequality by separating conditional measurement variation from between-unit variation. In addition, we prove that with n units and m measurements per unit, ReLU networks can attain an integrated mean squared error of order n^-1+(nm)^-2\beta/(2\beta+d) , up to logarithmic factors, over \beta -Hölder classes. We also derive a weighted oracle inequality for unequal cluster sizes and a rate for compositionally smooth functions. For pointwise ensemble inference, we give a projection central limit theorem and prove infinitesimal jackknife consistency under an explicit asymptotic linearity condition. Simulations and real data examples are provided to support our theoretical findings and practical implications.
[LG-97] Scalable Minimum-Volume Simplex Estimation with Non-asymptotic Analysis
链接: https://arxiv.org/abs/2609.25576
作者: Jun LI,Yanlong Guo,Zhaozhao Zeng
类目: Machine Learning (stat.ML); Machine Learning (cs.LG)
*备注:
Abstract:We study the estimation of a K -dimensional simplex from N i.i.d.\ points sampled uniformly from its interior; the observations are convex combinations of K+1 unknown prototypes. Existing polynomial-time estimators need cubic per-sample work or O(NK) storage and are impractical at N\sim 10^6 – 10^8 . We propose DeepMVSA, which re-expresses the minimum-volume principle in neural implicit form: a lightweight coordinate network generates the mixing weights and a triangular LU-type parameterization the dual simplex matrix, reducing the trainable-state memory to O(K^2) , independent of N , and the cost per data pass to O(NK^2) . We prove a non-asymptotic sample-complexity bound of the polynomial-time benchmark order for a localized surrogate estimator; an oracle inequality for every global minimizer of the neural objective, with volume-inflation control and an explicit shrinkage bias; a conditional end-to-end error budget separating statistical, approximation, optimization, and enclosure-residual terms on an explicit envelope event; and two-point lower bounds: at any noise level \sigma0 fixed independently of N , the N^-1/2 scaling is unimprovable in its N -exponent. Experiments with up to N=10^8 synthetic observations are consistent with the predicted accuracy and scaling, and feasibility on real scenes of \sim 10^7 pixels is demonstrated.
[LG-98] FAST-ML: A Hybrid Physics-Machine Learning Framework for Tropical Cyclone Intensity Forecasting
链接: https://arxiv.org/abs/2609.25505
作者: Shijie Xiao,Jonathan Lin,Thomas Ehrmann,Ali Sarhadi
类目: Atmospheric and Oceanic Physics (physics.ao-ph); Machine Learning (cs.LG)
*备注:
Abstract:Rapid intensification (RI) remains one of the most consequential and difficult aspects of tropical cyclone (TC) forecasting. Although full-physics numerical weather prediction models can represent the processes governing RI, resolving storm-environment interactions remains computationally expensive, while purely data-driven approaches often lack physical interpretability. We present FAST-ML, a hybrid framework that bridges data-driven efficiency with physical constraints. A physically informed dual-stream neural parameterization ingests 3D ERA5 fields to diagnose ventilation controls—environmental wind shear and mid-level entropy deficit. By optimizing these parameters end-to-end through a differentiable FAST intensity model, this architecture establishes a robust new paradigm for observation-driven parameter optimization, ensuring storm evolution remains strictly governed by thermodynamic principles. By better capturing the storm’s continuous intensity evolution, FAST-ML improves upon its physical baseline, reducing ensemble CRPS across forecast lead times, with a reduction of approximately 31% at 60 h and nearly halving the RI false alarm ratio without sacrificing detection skill. In a 100-member ensemble configuration, FAST-ML produces intensity forecasts comparable to FNV3 for selected storms under the evaluated input configurations. Furthermore, zero-shot tests on selected Eastern Pacific storms provide encouraging evidence of cross-basin transferability. FAST-ML provides a modular intensity forecasting framework that can be coupled with externally supplied storm tracks and environmental fields. It demonstrates that observation-driven parameter learning within physically constrained dynamics simultaneously enhances accuracy, interpretability, and computational efficiency.
[LG-99] Learnable Classifier-Free Guidance Null Embeddings for Enhanced Controllable Speech Synthesis INTERSPEECH2026
链接: https://arxiv.org/abs/2609.25411
作者: Biel Tura Vecino,Yoach Lacombe,Julian Weber,Zbigniew Łatka,Haitong Zhang,Logan Hart,Eren Gölge
类目: Audio and Speech Processing (eess.AS); Machine Learning (cs.LG)
*备注: Accepted paper at Interspeech 2026
Abstract:Classifier-free Guidance (CFG) is widely adopted in text-to-speech (TTS) systems to enhance generation quality and conditioning fidelity by interpolating between conditioned and unconditioned predictions. A common unconditional technique is to use an empty representation, in the form of a fixed null vector. In this work, we propose replacing this representation with a learnable unconditional embedding, optimized to represent a meaningful unconditional state. Objective and subjective evaluations demonstrate that learnable null embeddings consistently outperform fixed null embeddings across speaker similarity, speech stability, and expressiveness, while exhibiting greater robustness to larger guidance scales. We further show that learning a distinct unconditional embedding for each of the TTS conditioning modalities allows fine-grained control over speaker and text guidance, showcasing the trade-off between similarity and quality, and stability and expressiveness in the generated speech.
[LG-100] Penalized Nonreversible Langevin for Constrained Sampling
链接: https://arxiv.org/abs/2609.25381
作者: Pervez Ali,Weihao Dong,Xiaoyu Wang
类目: Machine Learning (stat.ML); Machine Learning (cs.LG); Probability (math.PR)
*备注: There are 68 pages and has 8 figures
Abstract:We propose penalized nonreversible Langevin algorithms for sampling from \pi(x)\propto e^-f(x)\mathbf 1_\mathcal C(x) , where \mathcal C\subset\mathbb R^d is a compact convex set. The algorithms combine a squared distance penalty with constant or compatible state dependent skew symmetric perturbations that preserve the penalized Gibbs distribution. For smooth, possibly nonconvex f , we derive nonasymptotic total variation bounds for the full gradient algorithm under a log Sobolev inequality. When unbiased stochastic gradients are available, we establish 2 -Wasserstein bounds under global contraction and Lipschitz conditions on the full drift in an adapted quadratic metric. For a fixed penalty parameter, the error relative to the penalized Gibbs distribution decays exponentially to an \mathcalO(\sqrt\eta) neighborhood, where \eta is the stepsize. We also bound the discrepancy between the penalized Gibbs distribution and the constrained target. In a two dimensional quadratic model, we establish nonreversible acceleration by tuning the skew perturbation to the curvature imbalance induced by penalization. With the target accuracy and smaller curvature fixed and initial Wasserstein distances uniformly bounded, tuning the skew perturbation improves the sufficient Euler iteration bound from linear to logarithmic in the curvature ratio. Numerical experiments evaluate the algorithms on constrained Bayesian regression, classification, neural networks, and truncated sampling, and examine the acceleration mechanism in a stochastic quadratic model.
[LG-101] Empirical Auditing of Edge-Private Graph Generators
链接: https://arxiv.org/abs/2609.25155
作者: Anum Fatima,Stratis Limnios,James Adams,Lukasz Szpruch,Carsten Maple,Gesine Reinert,Andrew Elliott
类目: Machine Learning (stat.ML); Machine Learning (cs.LG)
*备注:
Abstract:We empirically audit privacy leakage by testing whether outputs from edge-neighbouring inputs remain distinguishable, using statistically valid lower bounds on the privacy loss witnessed by our attacks. Our framework compares direct-edge, local-structural, and GNN-based attacks through the geometry surrounding a target edge. Experiments across two generators and two networks show that privacy leakage is both mechanism- and network-dependent, with learned representations revealing information not captured by conventional local statistics.
[LG-102] Variational objectives for amortized Bayesian inference in inverse problems: The role of posterior conditioning
链接: https://arxiv.org/abs/2609.25145
作者: Abhishek Srivastava,Arijit Hazra,Rajesh Dubbaku
类目: Machine Learning (stat.ML); Computational Engineering, Finance, and Science (cs.CE); Information Theory (cs.IT); Machine Learning (cs.LG); Numerical Analysis (math.NA)
*备注: 50 pages
Abstract:Variational autoencoders (VAEs) offer an efficient approach to amortized Bayesian inference for inverse problems, but posterior accuracy can depend strongly on the choice of variational regularization, particularly when the inverse problem contains weakly identified parameter directions. This study investigates three objectives: a reverse Kullback–Leibler formulation (VAE-KL), an asymmetric Jensen–Shannon formulation (VAE-JS), and a Jensen–Shannon–Wasserstein formulation (VAE-JSWA), which replaces the reverse Kullback–Leibler regularizer with the squared 2-Wasserstein distance while retaining forward-Kullback–Leibler posterior supervision. A full-covariance Gaussian encoder and a pre-trained physics-based surrogate are used for amortized posterior inference. A local linear–Gaussian analysis in the generalized Fisher basis is developed to characterize the variance-dependent gradients of the three objectives. The formulations are first evaluated using linear–Gaussian benchmarks with known posterior solutions and subsequently tested on nonlinear physics-based inverse problems, including an inverse problem governed by a linear ODE and two PDE-constrained problems. VAE-KL performs slightly better than the other formulations in the well-conditioned benchmark, where all three approaches yield comparable posterior approximations, whereas VAE-JSWA provides substantially lower posterior errors in the strongly ill-conditioned benchmark. The nonlinear physics-based problems exhibit a similar conditioning-dependent trend, with JS-based formulations providing greater benefit as posterior ill-conditioning increases. These results indicate that posterior conditioning is an important factor in selecting variational objectives and motivate geometry-adaptive variational inference for Bayesian inverse problems.
[LG-103] he Informational Content in Lepto-Variance and Its Relation to Higher Moments
链接: https://arxiv.org/abs/2609.25144
作者: Vassilis Polimenis
类目: Methodology (stat.ME); Machine Learning (cs.LG); Risk Management (q-fin.RM); Machine Learning (stat.ML)
*备注: 16 pages
Abstract:Lepto-regression is defined as the machine learning process of constructing a Regression Tree of a target feature on itself. It is a novel, model-free method potentially revealing information on important sample structure properties. But it is yet not clear what the informational content of lepto-variance is and how it is related to other well-known statistics of a sample. One significant finding is that 58% of the historical US stock return variability is 1-bit lepto-variance that can not be explained by any financial factor. The central question investigated in this paper is to use small normal N(0, 1) drawn samples to explore how the 1-bit sample lepto-variance and lepto-ratio relate to sample variance, skewness and excess kurtosis. Using a large sample simulation, the lepto ratio of a normal is found to converge to 36.3%. For smaller normally distributed simulated N(0, 1) samples, while lepto-variance itself is highly correlated to sample variance, lepto-variance as a fraction of total variance is highly correlated to excess kurtosis. Both lepto-variance and lepto-ratio are orthogonal to sample skew. Another finding is that while lepto-ratio is strongly correlated to lepto-variance it remains orthogonal to sample variance.
[LG-104] FREESIA: Covariance-Aware Posterior Transport for Expressive and Scalable Data Assimilation
链接: https://arxiv.org/abs/2609.25085
作者: Shiwei Ni,Yangwen Zhang,Hang Qi,Xiaofei Guan,Lili Ju
类目: Machine Learning (stat.ML); Machine Learning (cs.LG); Numerical Analysis (math.NA)
*备注:
Abstract:Data assimilation aims to infer the state of complex dynamical systems based on observational data. However, accurate inference of the multimodal posteriors induced by nonlinear or non-injective observation operators remains a key challenge under high-dimensional and sparse observation conditions. Ensemble filters scale to high dimensions but are confined by restrictive distributional assumptions, while training-free generative filters (e.g., EnSF, EnFF) alleviate this limitation but may introduce structural errors and hinder information propagation under sparse observations. To address these issues, we propose a training-free, asymptotically exact posterior transport method. Firstly, a covariance-aware posterior transport scheme is designed, which embeds the forecast cross-covariance into flow-based transport and accurately recovers unobserved states while preserving the non-Gaussian posterior structure. Furthermore, the method combines a tractable observation-adaptive proposal with posterior correction, ensuring accurate approximation of the nonlinear posterior distribution. Finally, we establish the corresponding posterior flow theory, from which the asymptotic exactness of the proposed method relative to finite-ensemble surrogates and the Wasserstein error bound are derived. Experiments on Double-Well, Lorenz-96, and Kolmogorov flow show that the proposed method captures complex posterior structure and remains accurate under sparse, nonlinear, and non-injective observations. In the sparse non-injective setting, it reduces RMSE by 56% relative to the best baseline.
[LG-105] End-to-End Quantum Semantic Communication with Variational Quantum Neural Networks
链接: https://arxiv.org/abs/2609.25044
作者: Melek Krichen,Nikhitha Nunavath,Riccardo Bassoli,Soumaya Cherkaoui
类目: Quantum Physics (quant-ph); Machine Learning (cs.LG)
*备注:
Abstract:This paper presents a quantum semantic communication (QSemCom) framework combining quantum machine learning (QML) and semantic communication (SemCom). Classical data are compressed into low-dimensional semantic representations, encoded and processed by a variational quantum transmitter, transmitted through a quantum channel, and processed by a trainable quantum receiver for classification. The framework considers a distributed quantum communication scenario in which quantum processing units (QPUs) exchange task-relevant semantic information through quantum links. While the general setting may involve multiple quantum nodes, this work focuses on the fundamental two-node case, with transmitter and receiver QPUs connected through a noisy quantum channel. Using MNIST, the framework is evaluated under ideal, bit-flip, depolarizing, and amplitude-damping channels. A baseline model is first trained over a perfect channel and evaluated under increasing noise without retraining. Receiver-side end-to-end training is then performed at fixed depolarizing-noise levels. The perfect-channel model achieves an accuracy of 0.9556 and an F1-score of 0.9551 . Results show channel-dependent performance degradation, while receiver training substantially restores task performance under moderate and high depolarizing noise. Moreover, task recovery does not require reconstruction of the transmitted density matrix, highlighting a distinction between physical-state recovery and semantic-feature recovery. These results demonstrate that a trainable quantum receiver can recover task-relevant semantic information from noise-distorted quantum states and maintain high classification performance.
[LG-106] What Does Chain-of-Thought Entropy Measure? A Channel Audit of Scaffolding Routing and Content
链接: https://arxiv.org/abs/2609.25039
作者: Marios Papamichalis,Regina Ruane
类目: Machine Learning (stat.ML); Information Theory (cs.IT); Machine Learning (cs.LG)
*备注:
Abstract:Entropy over chain-of-thought tokens decides which tokens receive the policy gradient, which get pruned, and whether a run has collapsed, yet each such statistic reads a next-token distribution mixing three choices: whether to emit connective scaffolding, which connective, and what the substantive continuation should be. Designating a scaffold vocabulary subset separates the three, exactly, for entropy, Kullback–Leibler divergence, and the first-order entropy velocity of a softmax policy. We prove the raw and content conventions disagree about which position is the larger fork on an explicit open region, and bound answer diversity by the content channel plus a leakage term a measured witness certifies. Across twenty-three configurations the scaffold side carries up to 41% of the raw high-entropy set; on a matched-tokenizer ladder, coupling changes only at the math-corpus step while the scaffold’s entropy share keeps growing through distillation; a closed-form forecast from one channel correlation tracks selection retention over a 54-point range to five points, unfitted. On compression, the content convention beats raw surprisal in every cell; an answer-leakage audit then corrects our own headline control: re-fed chains earn a quarter to a half of their accuracy from restated answers, and once stripped, no token scorer beats a random contiguous block.
[LG-107] From IceCube to IT-Sphere: A Hybrid Quantum-Classical GNN for Banking IT Root Cause Analysis
链接: https://arxiv.org/abs/2609.22822
作者: Antonio Greco,Riccardo Paoletti,Roberto Cappuccio,Mario Onorato
类目: Quantum Physics (quant-ph); Machine Learning (cs.LG)
*备注: 3 pages (cover page + 2-page extended abstract), 4 figures. Accepted for publication in the Proceedings of the 2026 IEEE International Conference on Quantum Computing and Engineering (QCE26), Toronto, ON, Canada, 13-18 September 2026; Poster Track, paper ID POS2-1600. Author’s accepted version; IEEE copyright notice on the first page
Abstract:We present Hybrid Quantum Root Cause Analysis (HQ-RCA), an industrially grounded workflow for root cause analysis in banking IT operations, built on a hybrid Quantum Graph Neural Network (QGNN): the classical backbone of DynEdge (the IceCube neutrino-reconstruction GNN, which we call standalone DynEdge), with its classification head replaced by a Variational Quantum Circuit (VQC). On 13 months of anonymised IT data (13k alarm clusters) from a major European bank, the hybrid QGNN matches standalone DynEdge – the strongest classical baseline – on F_1 , while standalone DynEdge leads the ranking metrics. A readout-sensitivity and layout-robustness study, analysed via Dimensional Expressivity Analysis (DEA), shows that the effective parameter dimensionality (rank) of the quantum observable has no measurable correlation with F_1 ; we therefore keep the simplest readout \langle Z_0\rangle (the Pauli- Z expectation on the first qubit), which in the deployed layout is rank-1, collapsing optimisation to a 1-D problem solvable by a gradient-free grid scan. Execution on IBM Heron r2 (no error mitigation) shows this gradient-free readout is executable on NISQ hardware after threshold recalibration.
附件下载
