本篇博文主要内容为 2026-08-21 从Arxiv.org论文网站获取的最新论文列表，自动更新，按照NLP、CV、ML、AI、IR、MA六个大方向区分。
说明：每日论文数据从Arxiv.org获取，每天早上12:30左右定时自动更新。
提示: 当天未及时更新，有可能是Arxiv当日未有新的论文发布，也有可能是脚本出错。尽可能会在当天修复。
目录
概览 (2026-08-21)
今日共更新511篇论文,其中:
- 自然语言处理共76篇(Computation and Language (cs.CL))
- 人工智能共146篇(Artificial Intelligence (cs.AI))
- 计算机视觉共86篇(Computer Vision and Pattern Recognition (cs.CV))
- 机器学习共127篇(Machine Learning (cs.LG))
- 多智能体系统共8篇(Multiagent Systems (cs.MA))
- 信息检索共15篇(Information Retrieval (cs.IR))
- 人机交互共19篇(Human-Computer Interaction (cs.HC))
多智能体系统
[MA-0] Multi-Agent Orchestration with the Common-Sense Reasoning Capabilities of LLM s for Autonomous Driving
【速读】：该论文旨在解决自动驾驶车辆在复杂且未见场景中面临感知与决策能力不足的问题，尤其针对传统强化学习（Reinforcement Learning, RL）和基于规则的方法在需要上下文推理时性能下降的局限性。同时，尽管大语言模型（Large Language Models, LLMs）在多模态信息理解与情境化推理方面表现出色，但其直接用于车辆控制时存在延迟高及生成幻觉（hallucination）的风险。为此，论文提出一种混合式框架，其核心在于通过一个协调器（orchestrator）协同使用经PPO训练的强化学习模块与比例-积分-微分（PID）控制策略，并将LLM的常识推理能力嵌入整个系统以增强情境理解。关键创新点在于：利用LLM进行迭代式推理，动态优化强化学习的奖励函数，从而提升系统在动态驾驶环境中的适应性与鲁棒性。实验在高度随机化的CARLA仿真环境中进行，涵盖多样化的环境与交通条件，结果表明该框架有效融合了LLM的上下文推理能力与传统自动驾驶方法的结构化控制与安全性，具备显著的应用潜力。
链接: https://arxiv.org/abs/2608.20129
作者: Mehdi Azarafza,Faezeh Pasandideh,Ali Ehteshami Bejnordi,Stefan Henkler,Achim Rettberg
机构: Hamm-Lippstadt University of Applied Sciences, Germany(哈姆-利普施塔特应用科学大学, 德国)
类目: Multiagent Systems (cs.MA); Computation and Language (cs.CL); Computer Vision and Pattern Recognition (cs.CV)
备注: 17 pages, 7 figures
Abstract:Autonomous vehicles require robust perception and decision-making capabilities to operate in diverse and unseen scenarios. While reinforcement learning and rule-based methods can provide effective control and safety mechanisms, their performance may degrade in situations requiring contextual reasoning. Large Language Models (LLMs) have demonstrated strong capabilities in understanding multimodal information and generating contextual reasoning, however, their use for direct vehicle control can introduce latency and hallucination risks. To address these limitations, a hybrid framework is proposed. This system uses an orchestrator to coordinate PPO-trained reinforcement learning and PID control, with LLM common-sense reasoning applied throughout the framework. LLM reasoning is further employed iteratively to refine the RL reward function for dynamic driving environments. The proposed framework is evaluated in highly randomized CARLA scenarios under diverse environmental and traffic conditions. The results demonstrate the potential of integrating LLM-based reasoning with conventional autonomous driving methods while retaining structured control and safety mechanism.
[MA-1] Reward-Guided Autoregressive Graph Generation for Efficient Multi-Agent Communication Topology Design ICONIP2026
【速读】：该论文旨在解决基于大语言模型（LLM）的多智能体系统（Multi-Agent Systems, MAS）在复杂推理任务中因智能体间通信与协作导致的高分词（token）消耗问题。现有自动拓扑设计方法ARG-Designer虽将拓扑生成建模为自回归图生成任务，但其训练目标未显式鼓励生成稀疏高效的拓扑结构，导致资源利用效率不足。为此，论文提出一种受奖励引导的自回归图生成方法（Reward-Guided Autoregressive Graph Generation, RGA-Designer），受人类反馈强化学习（Reinforcement Learning from Human Feedback, RLHF）启发，构建一个联合捕捉任务正确性与结构紧凑性的奖励模型，并以此作为反馈对预训练图生成器进行微调。该方案在保持与ARG-Designer相当的任务准确性的同时，平均降低20.5%的分词消耗，其核心创新在于通过奖励机制显式优化拓扑结构的稀疏性与高效性。
链接: https://arxiv.org/abs/2608.20099
作者: Poomphob Suwannapichat,Boonyarit Changaival,Caesar Wu,Pascal Bouvry
机构: University of Luxembourg(卢森堡大学); King Mongkut’s University of Technology Thonburi(泰国曼谷皇家理工大学)
类目: Multiagent Systems (cs.MA); Computation and Language (cs.CL); Machine Learning (cs.LG)
备注: Full version of extended abstract accepted at ICONIP 2026 (poster)
Abstract:LLM-based Multi-Agent Systems (MAS) achieve strong performance on complex reasoning tasks by coordinating multiple agents, but at the cost of substantial token consumption. Recent work on automatic topology design, ARG-Designer, has reframed this problem as autoregressive graph generation. However, its training objective provides no explicit incentive for the model to generate sparse and efficient topologies. We address this limitation by introducing a Reward-Guided Autoregressive Graph Generation (RGA-Designer) inspired by Reinforcement Learning from Human Feedback (RLHF). We train a reward model that jointly captures task correctness and structural compactness, and then fine-tune the pretrained graph generator using the reward model as feedback. Our method preserves task accuracy at the level of ARG-Designer while reducing token consumption by an average of 20.5%.
[MA-2] What You Cant See Is What You Learn: Restricted Evidence Visibility Favors Compositional Generalization in Shared-Genome Language-Model Societies
【速读】：该论文旨在解决多模块系统在梯度训练过程中因模块间全输入可见性导致的泛化能力不足问题，探究受限证据可见性（restricted evidence visibility）是否能够影响模型发现更优解。其核心解决方案在于通过设计一种通信受限的架构：四单元社会仅通过两个固定宽度的连续向量进行通信，并强制使用注意力掩码限制信息流动，从而实现对输入信息的局部可见性控制。实验表明，在自然语言函数组合任务中，受限可见性的社会在9/10的配对中优于全局可见的对照组，且在深度为三时仍保持0.558的性能优势，证明了受限通信能显著提升泛化能力。进一步分析显示，受限系统具备可复用的价值索引接口特性，同值数据包移植后行为保持稳定（0.94–1.00），而破坏性干预则导致性能崩溃，反事实数据包可引导输出至数学预测结果，表明系统内部存在可解释的推理机制。尽管受限方案在多数指标上表现优异，但因整体中位数深度三准确率（0.6988）未达预设阈值（0.70），完整实验协议仍被判定失败。此外，唯一高性能全局模型虽依赖通信，但其同值数据包不具备跨轮次可交换性，凸显受限可见性在促进可重用、通用接口构建方面的独特价值。
链接: https://arxiv.org/abs/2608.20054
作者: Narcis Marincat
机构: 未知
类目: Artificial Intelligence (cs.AI); Machine Learning (cs.LG); Multiagent Systems (cs.MA)
备注: 16 pages, 3 figures, 5 tables. Code and evaluation records: this https URL ; checkpoints: this https URL
Abstract:Multi-module systems often expose every module to the full input. We test whether restricting evidence visibility changes which solutions gradient-based training discovers. Four-cell societies share one frozen pretrained language model and one low-rank adapter, communicating only through two model-width continuous vectors in a fixed relay. On a prospectively sealed natural-language function-composition task, we train ten matched restricted/global pairs sharing initialization bytes, training order, token layout, parameters, and computation; only the attention mask differs. Restricted societies outperform their globally visible twins by at least 20 points at both depths in 9 of 10 pairs, with median paired advantages of 0.7648 and 0.6050. Cutting communication reduces every restricted society to chance, and the depth-three advantage remains 0.558 on programs whose composite function never appeared in training. Across six audited restricted societies, same-value packet transplants preserve behavior at 0.94-1.00 across all tested interfaces; destructive interventions collapse performance; and counterfactual packets redirect outputs toward the mathematically predicted answer. The sole high-performing global model also requires communication, but its same-value packets are not interchangeable across episodes. Restricted visibility is thus not necessary for composition; under this protocol it substantially increases the probability of a generalizing relay and favors a reusable, value-indexed interface. The complete preregistered battery nevertheless formally fails because restricted-arm median depth-three accuracy is 0.6988, below the 0.70 floor. An earlier qualification cohort likewise yielded 0/10 complete passes: one model met every task-performance gate, but all ten failed ordinary-language preservation, confining the system to explicitly task-gated use.
[MA-3] Distilling Aggregate Mobility Statistics into a Language Model Policy for Post-Event Crowd Simulation
【速读】：该论文旨在解决行人模拟器中个体行为规则难以准确建模的问题，其核心挑战在于隐私限制导致可用数据仅限于聚合统计信息（如区域设备数量和起讫点（Origin-Destination, OD）流量），而缺乏个体轨迹数据。此类聚合数据无法唯一确定个体行为决策，因为存在多种不同的行为组合可产生相同的统计结果。为此，论文提出通过微调语言模型驱动的群体代理（crowd agent），使其模拟出的群体目的地分布与实际观测到的目的地组成一致。关键解决方案是：从OD流中读取目标目的地分布，并利用迭代比例调整（Iterative Proportional Fitting, IPF）对模型自身的目的地分布进行重加权，以匹配真实分布。由于微调过程会放大主导目的地类别，因此进一步采用低秩适配器（low-rank adapter）在重新采样至校正后训练分布的轨迹上进行拟合，从而在微调后仍能实现目标分布。在两次棒球比赛的移动网络统计数据上的实验表明，经微调后的代理无需推理时修正，即可将目的地占比误差降低25%，同时各类策略下的网格相关性保持稳定。
链接: https://arxiv.org/abs/2608.19778
作者: Tatsuya Amano,Hirozumi Yamaguchi
机构: The University of Osaka(大阪大学); RIKEN Center for Computational Science(理化学研究所计算科学中心)
类目: Multiagent Systems (cs.MA); Artificial Intelligence (cs.AI)
备注: 4 pages, 3 figures. Accepted as a short paper at ACM SIGSPATIAL 2026
Abstract:Pedestrian simulators need a behaviour rule for every agent, but privacy usually limits the data for setting one to aggregate statistics, namely zone-level device counts and origin-to-destination (OD) flows, with no individual trajectories. Such aggregates under-determine individual behaviour, because many different sets of decisions reproduce the same counts. We fine-tune a language model crowd agent so that the simulated population matches the observed destination composition, the fraction of the departing crowd heading to each point of interest. We read this target from the OD flow and reweight the model’s own destination distribution onto it by iterative proportional fitting. Because fine-tuning inflates the dominant destination class, we fit the low-rank adapter to trajectories resampled to a corrected training composition that reaches the target after this inflation. On mobile network counts from two baseball games the fine-tuned agent runs without inference-time correction, cutting the destination-share error by 25%, while the grid correlation remains similar across policies.
[MA-4] An Evidence-Grounded Multi-Agent System for High-Level Bio-Robot Design
【速读】：该论文旨在解决生成式人工智能在合成生物学领域中对生物机器人（bio-robot）系统进行高精度、可追溯的模块化设计支持的问题，尤其针对基于基因回路编程的活体或类生物混合系统的复杂设计需求。其核心挑战在于如何将应用需求精准转化为传感、逻辑、记忆、输出、组装、宿主与隔离等模块，并确保每个组件的选择均基于可验证的生物元件数据和实证依据。解决方案的关键在于构建一个基于Qwen3.5-27B的离线多智能体系统——micro_biorobot_agent，该系统通过整合包含23,762条记录的综合知识库（涵盖生物元件、实测组合、文献支持关系及执行证据），实现了从需求分析到候选装配、冲突检测、局部修复、独立评审与验证的全流程自动化。系统引入确定性输出校验机制，有效纠正虚假遗漏、无支持的元件提及及溯源错误；在两组各50个查询的评估集上，平均得分分别达到7.35和8.04，显著优于其他七种对比系统，在场景设计任务中领先第二名2.23分。50个查询的配对消融实验表明，源追踪检查使虚假遗漏事件从15例降至3例（降低80%），同时提升溯源准确性0.75分，凸显了其在保障设计可信度方面的关键作用。本研究报告的是基于Qwen3.5的v1版本系统，重点评估高层次设计报告的质量，而非实验验证的电路实现。
链接: https://arxiv.org/abs/2608.19699
作者: Yujun Chen,Tianle Li,Jiayu Chen,Zhen Yin
机构: 未知
类目: Multiagent Systems (cs.MA)
备注: 15 pages, 2 figures, 9 tables, and 4 algorithms
Abstract:In this paper, a bio-robot is an engineered living or biohybrid system in which living cells perform one or more core functions, such as sensing, information processing, actuation or output. We focus on systems whose cell-based functions are programmed by genetic circuits; physical movement is optional. Designing such a system requires translating application requirements into sensing, logic or memory, output, assembly, host and containment modules, while grounding each choice in traceable parts and evidence. We present micro_biorobot_agent, an offline multi-agent system built on Qwen3.5-27B. The system combines requirement analysis, module-specific retrieval, candidate assembly, conflict checking, local repair, independent review and validation over an integrated library of 23,762 records covering biological parts, measured combinations, literature-supported relationships and actuation evidence. Deterministic output checks align the final report with the retrieved part set and correct false gaps, unsupported part mentions and source-tracking errors. On two author-developed evaluation sets of 50 queries each, the system obtains mean overall scores of 7.35 and 8.04, the highest among the seven evaluated systems; on Scenario Design it exceeds the runner-up by 2.23 points. A 50-query paired ablation shows that the source-tracking check reduces false-gap incidents from 15 to 3, an 80% reduction, and increases source accuracy by 0.75 points. This paper reports the Qwen3.5-based v1 system and evaluates high-level design reports rather than experimentally validated circuits.
[MA-5] When Do LLM Agents Help? Deadline-Aware Mixed-Criticality Task Scheduling at the Autonomous-Vehicle Edge
【速读】：该论文旨在解决自动驾驶车辆在移动边缘计算（MEC）环境中对低时延感知任务进行调度时面临的混合关键性（mixed-criticality）挑战，尤其关注安全关键型任务（time-critical, TC）在高负载与非平稳场景下难以保障完成率的问题。其核心问题是：在异构MEC服务器上，如何在保证时间敏感任务优先执行的前提下，实现高效且可解释的动态资源调度，同时评估大型语言模型（LLM）作为智能控制层是否能超越传统启发式方法。解决方案的关键在于构建一个基于窗口化契约网拍卖机制的强启发式算法，该算法通过“最早截止时间优先”（earliest deadline first）排序和“最早完成时间分配”策略，在60个不同拓扑实例上实现了0.902的TC任务完成率，显著优于15种基线方法（Holm校正后p < 0.001），并接近CP-SAT上界（达0.87）。进一步引入多智能体LLM控制层后发现，当系统负载平稳时，LLM带来的额外收益微乎其微；但当出现中段运行中的安全关键任务突发（non-stationarity）时，LLM凭借在线自适应能力显著优于静态启发式与强化学习基线。因此，LLM控制层的价值仅在非平稳环境下显现，其优势来源于对批处理窗口长度与时间关键任务优先排序等基本因子的动态优化，而非复杂的拍卖机制或局部策略。研究还报告了控制平面的延迟与决策推理过程，并开源全部代码与种子实例。
链接: https://arxiv.org/abs/2608.19557
作者: Reza Zakerian
机构: Westcliff University (西克利夫大学)
类目: Distributed, Parallel, and Cluster Computing (cs.DC); Multiagent Systems (cs.MA)
备注: 8 pages, 5 figures, and 4 tables
Abstract:Autonomous vehicles offload latency-sensitive perception tasks to nearby mobile edge computing (MEC) servers, where a missed safety-critical task is unsafe rather than merely degraded. Large language models (LLMs) are increasingly proposed as adaptive, explainable schedulers, yet evidence of when they help is scarce. We study deadline-aware, mixed-criticality scheduling on heterogeneous MEC servers, where time-critical (TC) tasks must be protected at a controlled cost to best-effort traffic, and ask whether a multi-agent LLM control layer improves on a strong heuristic. We answer in two steps. First we build the heuristic: a windowed contract-net auction that orders each admission window time-critical-first by earliest deadline and places tasks by earliest-finish-time. Across 60 instances on three topologies and 15 baselines under an identical online constraint, it attains a TC completion rate of 0.902, above every baseline (Holm-corrected p 0.001; best baseline 0.838) and at 0.87 of a CP-SAT upper bound. Second, we add the LLM control plane. A controlled decomposition traces the scheduler’s advantage to two ordinary factors, the batching horizon and time-critical-first ordering; the auction, the per-window LLM policy, and online adaptation add nothing while the load is stationary, where the heuristic is already near-optimal. Under a mid-run surge of safety-critical tasks the picture changes, and the LLM control plane gains significantly over both the static heuristic and the bandit. LLM orchestration therefore earns its cost only when non-stationarity opens headroom a fixed policy cannot use. We report control-plane latency and rationale, and release all code and seeded instances.
[MA-6] he Evaluation Context Protocol (ECP): A Portable Contract for AI Agent Evaluation
【速读】：该论文旨在解决当前人工智能评估体系在面对自主智能体（AI agents）时的系统性缺陷，尤其是传统评测范式无法有效衡量复杂代理架构的实际运行可靠性与真实能力。其核心问题在于：现有评估方法存在基准漏洞（benchmark exploitation）、“自信错误”（confidently wrong）现象以及理论能力与实际操作可靠性之间的脱节。为应对这一挑战，论文提出关键解决方案——评估上下文协议（Evaluation Context Protocol, ECP），这是一个早期阶段、厂商中立的可移植评估契约层，旨在统一不同框架间对智能体行为的可观测性标准。ECP通过定义一个基于JSON-RPC的小型接口，使智能体能够暴露其用户可见输出、所调用工具及受评者安全的审计上下文，从而支持跨框架与持续集成系统的程序化一致性验证。该方案提供了一个开源参考实现，兼容LangChain、LlamaIndex、CrewAI和PydanticAI等主流开发框架，并针对近期文献中记录的失效模式进行设计校准。值得注意的是，ECP目前仍处于演进阶段，其评估维度、方法集与评分器家族均预期随实际应用迭代优化，其广泛采纳所需的实证验证亦被列为未来工作。
链接: https://arxiv.org/abs/2608.19263
作者: Aniket Wattamwar,Manav Anandani,Mrunal Kakirwar
机构: 未知
类目: oftware Engineering (cs.SE); Multiagent Systems (cs.MA)
备注: 14 pages, 4 tables, 4 figures, Code available at this https URL
Abstract:The evolution of artificial intelligence has necessitated a fundamental shift from evaluating isolated Large Language Models (LLMs) to assessing autonomous agentic architectures. This paper explores the critical methodologies for evaluating AI agents and the essential role of advanced observability infrastructure. We analyze the architectural components of agents and identify the severe limitations of current evaluation paradigms, including benchmark exploitation, the “confidently wrong” phenomenon, and the discrepancy between theoretical capability and operational reliability. To begin addressing the fragmentation in current evaluation infrastructure, this paper proposes the Evaluation Context Protocol (ECP), an early-stage, vendor-neutral framework intended to act as a portable evaluation contract layer for agentic systems. In its current form ECP defines a small JSON-RPC interface over which an agent exposes its user-visible output, the tool calls it made, and evaluator-safe audit context, and against which programmatic checks can be run uniformly across frameworks and continuous integration systems. We describe an open-source reference implementation that includes adapters for LangChain, LlamaIndex, CrewAI, and PydanticAI, and we situate the design against failure modes documented in the recent literature. ECP is presented as work in progress rather than a finished standard: the evaluation surface, method set, and grader families are all expected to change as the protocol is exercised against more systems, and the empirical validation required to justify adoption is outlined as future work.
[MA-7] Hallucination as a Feature not a Defect: Evaluating a multi-agent architecture to transform speculative language-model outputs into testable scientific hypotheses
【速读】：该论文旨在解决当前大语言模型（Large Language Models, LLMs）在对齐过程中过度抑制幻觉（hallucination）所导致的创造性受限问题，即模型因优先追求事实准确性而出现语义过拟合（semantic overfitting）与多样性崩溃（diversity collapse），从而阻碍了探索性研究与开发（Speculative Research and Development, RD）。其解决方案的关键在于构建一个基于Rust的多智能体协同架构，通过类比叙事性“白日梦”（narrative daydreaming）与执行控制（executive control）之间的动态张力，实现认知功能的分工：高熵生成智能体负责产生多样化、非结构化的创意假设，而基于网络检索的评估智能体则提供实证约束与外部验证；两者之间由一个低熵语义瓶颈（semantic bottleneck）进行信息过滤，以降低冗余与噪声。实验表明，该系统在物理与社会科学领域可生成多样且具备可行性评级的假设，尤其在面对强物理、实证或制度约束时表现出显著优势。研究还揭示，单纯直接提示（direct prompting）表现最差，而全系统并未普遍优于简单自我反思（self-reflection），说明不同架构在原创性、可行性、多样性与实证基础之间的权衡机制各不相同。核心结论是：幻觉本身并非价值所在，唯有在架构化约束、实证锚定与显式评估的共同作用下，推测性生成才具有实际意义。
链接: https://arxiv.org/abs/2608.19206
作者: Nicolas Rodriguez-Alvarez(IES Parquesol, Valladolid, Spain)
机构: IES Parquesol(帕尔奎索尔中学), Valladolid, Castilla y León, Spain(西班牙卡斯蒂利亚-莱昂自治区瓦拉多利德)
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI); Multiagent Systems (cs.MA)
备注: 25 pages. Bilingual: full English version followed by the complete Spanish version. Includes an exploratory paired baseline and ablation study (6 conditions). Code and data: this https URL
Abstract:Contemporary Large Language Models (LLMs) are increasingly aligned to suppress hallucinations, prioritizing factual retrieval over combinatorial creativity. While crucial for mitigating misinformation, this alignment may also restrict speculative Research and Development (RD) by encouraging what this work operationally treats as semantic overfitting and diversity collapse. In this paper, we propose a Rust-based multi-agent orchestration that uses the contrast between narrative daydreaming and executive control as a functional analogy, not as a neurocognitive claim. The system instigates an Epistemological Friction loop between a high-entropy generating agent and a web-grounded evaluating agent, mediated by a low-entropy semantic bottleneck intended to reduce noise and repetition. Initial experiments generated diverse, viability-rated hypotheses across physical and social-science domains. We additionally report an exploratory paired baseline and ablation study comparing the full system against direct prompting, self-reflection, removal of the semantic filter, removal of search grounding, and removal of lateral lenses. The results place direct prompting among the weakest conditions across most observed metrics, but they do not show a general superiority of the full system over simple self-reflection. Instead, they suggest that each architecture shifts the balance between originality, feasibility, diversity, and empirical grounding in different ways, and that the full system provides its main advantages when hypotheses must survive strong physical, empirical, or institutional constraints. These findings do not show that hallucination is useful in isolation; they suggest that speculative generation gains value only when constrained by architecture, empirical grounding, and explicit evaluation.
自然语言处理
[NLP-0] ConceptGuard: Benchmarking Context-Sensitive Unlearning in Large Language Models NEURIPS
【速读】： 该论文旨在解决大语言模型（Large Language Models, LLMs）在实际应用中对有害或敏感知识进行选择性删除（即“遗忘”，unlearning）时，现有方法与评估基准无法全面衡量其有效性的核心问题。当前方法通常依赖于独立事实构成的遗忘集与保留集，并以简单的事实召回作为评价标准，这忽略了真实场景下关键需求：在消除有害行为的同时，完整保留有益且无害的知识使用方式。为此，论文提出解决方案的关键在于引入“双重用途概念”（dual-use concepts）——即同一概念可在有害与良性语境中被使用。基于此，构建了名为ConceptGuard的新基准，其中遗忘集与保留集在概念层面具有明确的互补性，使评估能够聚焦于概念层级而非零散事实。该基准强调意图敏感性，目标是最大化上下文区分度，从而推动更安全的行为表现。实验表明，现有遗忘技术在此设定下表现不佳，存在显著的遗忘-效用权衡、上下文敏感性提升有限以及概念级控制不一致等问题，揭示出当前方法在实现真正意义上概念性、完整性遗忘方面仍存在严重不足。该研究为设计更符合现实安全需求的遗忘机制提供了重要方向。
链接: https://arxiv.org/abs/2608.20338
作者: Sahil Kale,Ian Harris
机构: Pune Institute of Computer Technology(普奈计算机技术学院); University of California, Irvine(加州大学欧文分校)
类目: Computation and Language (cs.CL)
备注: Submitted to NeurIPS ED Track 2026; 17 pages, 9 figures
Abstract:Large Language Models (LLMs) increasingly require selective removal of harmful or sensitive knowledge, called unlearning, yet existing methods and benchmarks fail to evaluate this capability completely. Current approaches rely on disjoint forget and retain sets composed of independent facts, and measure success using simple and direct factual recall. This framing fails to capture a key requirement of unlearning, namely the ability to eliminate harmful behaviors while preserving benign and beneficial knowledge. We argue that effective unlearning must operate at the level of concepts, ensuring complete removal of unsafe applications while maintaining their correct and useful usage, thereby achieving conceptually meaningful and complete unlearning. To better evaluate unlearning techniques from such a practical viewpoint, we introduce the notion of dual-use concepts: concepts that can be used in both harmful and benign contexts. Building on these concepts, we construct a benchmark called ConceptGuard where forget and retain sets are explicitly complementary in concept usage. Our benchmark uniquely enables unlearning to be explored and gauged at the level of concepts, instead of sparse facts, and evaluation is intent-sensitive with the goal of maximizing contextual separation to promote safer behavior. We demonstrate that current unlearning techniques perform poorly under this setting, showing weak contextual separation alongside poor performance in ROUGE and concept-level metrics. Our results reveal strong forgetting-utility trade-offs, limited gains in contextual sensitivity, and poor consistency in concept-level control across methods, and provide ideas for unlearning approaches that better align with real-world safety requirements. Our dataset is publicly available.
[NLP-1] G-CARL: Grounded Checklist-Aligned Reward Learning for Patient-Oriented Medical Report Interpretation
【速读】： 该论文旨在解决患者对医学报告个性化解读的迫切需求，即在保证医学事实准确性（evidence-grounded medical factuality）的同时，实现基于用户上下文的可理解性与沟通适配性（context-dependent patient communication）。现有医学视觉-语言任务难以同时满足这两类目标，因其在可验证性上存在根本差异，且二者紧密耦合，传统监督微调与整体强化学习方法难以协同优化。为此，论文提出G-CARL框架——一种基于事实核查（grounded）、检查清单对齐（checklist-aligned）的强化学习方法，通过多源检索实现原子命题的逐项验证，并结合上下文感知、实例特异的加权检查清单，对响应覆盖度进行结构化监督，从而在不约束生成多样性前提下，同步提升事实性、用户需求满足度与表达质量。研究还构建了真实世界场景下的PMRI基准数据集MMedReport及由临床医生设计的三维评估体系，实验表明G-CARL在整体质量、命题级精确率和检查清单召回率方面均优于现有后训练基线，临床医生的成对偏好评估进一步证实其生成结果更具准确性并更契合患者实际需求。
链接: https://arxiv.org/abs/2608.20331
作者: Shiao Xie,Siyu Chen,Jianwei Lv,Bo Yuan,Yujin Wang,Xiandong Li
机构: 未知
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI); Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Personalized interpretation of medical reports has emerged as an increasingly important need among patients. Addressing this need requires both evidence-grounded medical factuality and context-dependent patient communication, yet existing medical vision-language tasks do not adequately capture these dual requirements. To bridge this gap, we introduce Patient-oriented Medical Report Interpretation (PMRI), a novel open-ended multimodal generation task that requires models to explain medical reports in accurate and accessible language based on a user’s query and dialogue history. These two objectives differ fundamentally in their verifiability, yet remain tightly coupled, making them difficult to optimize jointly under conventional supervised fine-tuning and holistic reinforcement learning paradigms. To address this challenge, we propose G-CARL, a grounded, checklist-aligned reinforcement learning framework that combines multi-source retrieval for atomic claim verification with context-aware, instance-specific weighted checklists for response coverage, providing structured supervision for factuality, user-demand satisfaction, and expression quality without constraining response diversity. We further construct MMedReport, a real-world PMRI benchmark, along with a clinician-designed three-dimensional evaluation protocol. Extensive experiments demonstrate that G-CARL consistently outperforms existing post-training baselines in overall quality, claim-level precision, and checklist recall. Pairwise preference evaluation by clinicians further confirms that G-CARL produces interpretations that are more accurate and better aligned with patient needs.
[NLP-2] An Agent ic Approach for Active Data Collection Travel Behavior Modeling and Weather-Sensitive Demand Prediction
【速读】： 该论文旨在解决旅行行为研究中数字数据采集与预测建模环节长期分离的问题，导致研究流程割裂、信息利用不充分。其核心解决方案是提出一种三智能体（three-agent）协同工作流，集成对话式数据采集、结构化数据处理与行为预测三大模块。关键在于通过聊天机器人实施基于图像增强的陈述偏好调查，结合多模型对比分析，系统评估不同配置下大语言模型（LLM）在零样本（zero-shot）、角色扮演（persona）、少量示例（few-shot）及视觉增强（vision-based）等条件下的预测性能。研究表明，尽管随机森林模型达到69.6%的五分类准确率，但最优文本型零样本LLM亦达69.9%，无需任务特定微调；引入习惯性出行信息可显著提升模型表现，专家型表述优于角色扮演，而视觉上下文信息使最佳多模态配置准确率提升至71.5%，验证了视觉情境对行为预测的增量价值。整体框架实现了从数据采集到预测建模的端到端可审计协同，为融合生成式人工智能与传统行为建模提供了可复现的技术路径。
链接: https://arxiv.org/abs/2608.20320
作者: Narges Ahmadi(1),Yubo Jiao(1),Jônatas Augusto Manzolli(1),Jiangbo Yu(1),Luis Miranda-Moreno(1) ((1) McGill University)
机构: 未知
类目: Artificial Intelligence (cs.AI); Computation and Language (cs.CL)
备注:
Abstract:Travel behavior research increasingly combines digital data collection with predictive modeling, yet these stages are often developed and evaluated separately. This study proposes a three-agent workflow integrating conversational data collection, structured data processing, and behavioral prediction. A chatbot-administered, image-augmented stated-preference survey collected mode choices from student commuters across five predefined weather scenarios, yielding 454 respondent-scenario observations. Weather-related associations were analyzed using a multinomial logit model, while logistic regression and random forest provided machine-learning benchmarks. Nine locally deployed large language models (LLMs), ranging from 2 to 35 billion parameters, were evaluated across four zero-shot prompt-and-context conditions and extended through persona, few-shot, and vision-based configurations. Random forest achieved 69.6% five-class accuracy, while the best text-only zero-shot LLM reached 69.9% without task-specific fitting. Habitual travel information produced the most consistent gains, Expert framing generally outperformed Role-Play, and persona information was most useful when habitual travel information was unavailable. Few-shot prompting improved prediction for several models, with gains stabilizing after a small number of examples. Using the same weather images shown to respondents, the best vision-based configuration reached 71.5% five-class accuracy, indicating that visual context may provide additional predictive information for selected models. Overall, the study shows how conversational surveys, structured data processing, conventional behavioral modeling, machine learning, and multimodal LLM prediction can be coordinated within an auditable multi-agent workflow.
[NLP-3] Inducing Task Models from Computer-Use Traces
【速读】： 该论文旨在解决从自然状态下被动记录的计算机使用轨迹（如屏幕截图、鼠标或键盘操作）中自动构建可符号化、可审计且可复用的任务模型这一挑战。现有方法受限于预设任务或单一工作流假设，仅能生成步骤级摘要，无法捕捉多线程、交错目标的真实工作场景。其核心问题在于如何在无监督条件下识别隐藏任务并构建具有层次化目标分解与控制流程结构的结构化任务模型。本文提出任务模型诱导（Task Model Induction, TMI），其关键创新在于：(i) 从非约束性轨迹中发现潜在任务，有效分离并发活动；(ii) 针对每个潜在任务，联合构建递归目标分解的层次化目标模型与组织执行过程的控制流程序模型。在受控的人类与代理轨迹上，TMI实现了与真实任务分组0.974的高一致性，并重构了74.9%的观测执行步骤，显著优于最强基线；此外，基于TMI生成的技能在保留任务上使准确率提升30.0%，验证了其在实际应用中的有效性。
链接: https://arxiv.org/abs/2608.20319
作者: Yucheng Jiang,Zora Zhiruo Wang,Ruishi Chen,Diyi Yang
机构: Stanford University (斯坦福大学); Carnegie Mellon University (卡内基梅隆大学)
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
备注:
Abstract:Naturalistic computer-use traces, passively recorded screenshots and mouse or keyboard actions, are a valuable resource for deriving symbolic, auditable, and reusable models of how everyday work is done. Such models matter as computer-use agents enter real work, where agents need to learn how tasks are actually performed, and organizations need to audit and reuse that knowledge. However, inducing such task models is challenging, as activity is observed only as low-level events and real-world work is multi-threaded with interleaved goals. Existing methods assume a given task or a single workflow, and produce step-level summaries rather than structured task models. We introduce Task Model Induction (TMI), which (i) discovers the latent tasks in an unconstrained trace, disentangling concurrent activity, and (ii) for each latent task, induces a task model pairing a hierarchical objective model of recursive goal decomposition with a procedure model of the control flow that organized the execution. Intrinsically, on controlled human and agent trajectories, TMI recovers interleaved tasks with 0.974 agreement against ground-truth groupings and reconstructs 74.9% of the observed execution steps, far more than the strongest workflow induction baseline. Extrinsically, skills derived from TMI’s task models improve held-out task accuracy by 30.0% over the strongest baseline.
[NLP-4] AI4AI-Bench: Benchmarking LLM Agents in Algorithmic Design for Recursive Self-Improvement
【速读】： 该论文旨在解决生成式人工智能（Generative AI）系统是否能够实现递归自我改进（Recursive Self-Improvement, RSI）这一核心问题，即一个智能体能否优化其自身训练算法，从而在后续迭代中提升整体性能。其关键挑战在于：现有基准测试无法有效隔离对训练算法设计能力的评估——多数现有评测依赖数据收集或超参数调优，而未考察对训练过程执行逻辑的根本性修改。为此，研究提出AI4AI-Bench，一个包含10个冻结的研究仓库、覆盖10类训练算法的任务基准。每个任务中，智能体在单个B300实例上运行4小时，重写训练算法；随后代码从头重新执行最多12小时，由隐藏的固定评估器进行评分，与原始算法在相同流程下对比。由于10个评价指标不可通约，所有任务均映射至统一尺度：0表示无信息模型，0.1为原始算法表现，1.0为目标最优。在6种系统、29种配置下，平均得分仅为0.166，最佳系统达到0.250，表明当前最强系统仍未能接近理论最优与已有算法之间的差距。分析显示，绝大多数提交未改变模型学习机制，仅少数改变学习方式的样本平均得分为0.226，显著高于其余样本的0.126。进一步发现，更多推理投入显著提升了尝试修改训练机制的意愿，使此类提交比例从8%上升至64%，平均得分也从0.094提升至0.196。研究公开全部任务、评估器及评分结果，以支持未来系统演进下的可重复测量。
链接: https://arxiv.org/abs/2608.20318
作者: Yizhe Chi,Wenyi Li,Deyao Hong,Xiaoqiu Wang,Mingju Gao,Kaisen Yang,Bingxiang He,Youjie Zheng,Calvin Xiao,Qinhuai Na
机构: Navers Lab(奈沃实验室); Einsia.AI(艾因赛); Tsinghua University(清华大学)
类目: Artificial Intelligence (cs.AI); Computation and Language (cs.CL); Machine Learning (cs.LG)
备注:
Abstract:Recursive self-improvement (RSI) asks whether an AI system can improve the process that produces AI systems, so that the next system inherits the improvement. That process is the training algorithm: a better objective or update rule improves the compute\mbox-capability exchange rate for every subsequent run, including the one that produces the next agent. Whether RSI is feasible therefore turns on whether an agent can design training algorithms. No benchmark isolates that ability: existing suites are won by collecting data or by tuning hyperparameters, and none tells a change to how a run is executed apart from a change to how the model learns. We present AI4AI\mbox-Bench, 10 frozen research repositories spanning 10 training algorithm families. In each task, an agent has 4 hours on one B300 to rewrite the training algorithm; its code is then rerun from scratch for up to 12 hours and scored by a fixed evaluator hidden from the agent, against the repository’s original algorithm under the same procedure. Because the 10 metrics are incommensurable, every task is mapped onto one scale on which 0 is an uninformative model, 0.1 is the algorithm the repository ships, and 1.0 is the task optimum. Across 29 configurations of 6 systems on all 10 tasks the mean score is 0.166 , and the best system reaches 0.250 : even the strongest closes under a fifth of the distance between the algorithm that was already there and the optimum. The submissions show where that distance went: most never change how the model learns at all, and the minority that do average 0.226 against 0.126 for the rest. More reasoning effort mostly buys the willingness to go there, taking that minority from 8% of submissions to 64% and the mean score from 0.094 to 0.196 . We release the task suite, the evaluators and every scored submission, so that the measurement can be repeated as these systems change.
[NLP-5] Phantom Gains: Auditing Self-Improvement Against a Measured Null
【速读】： 该论文旨在解决当前语言模型自我改进评估中因依赖噪声估计差异而导致的测量偏差问题，尤其关注在自训练（self-training）过程中对个体问题性能变化的误判。其核心挑战在于，现有评估方法常通过对比不同轮次的模型表现来判断性能增益或损失，但这种差分操作极易受采样噪声和推理过程中的非理想因素干扰，从而产生虚假的“性能提升”或“退化”结论。解决方案的关键在于提出一种基于多臂实验设计的过渡级审计（transition-level auditing）框架：为每个统计量构建独立、可复现的零假设（null hypothesis），该零假设不依赖额外实验，而是利用已有基准模型的重复实验数据，通过在错误发现率（FDR）控制下的逐问题精确检验（exact test）实现。这种方法有效避免了传统指标如贪婪解码生成的伪能力变化、扩增统计量的误导性解释以及阈值修复在复制中的失效等问题。实证结果显示，外部知识蒸馏显著提升了基础模型难以触及的问题，而三种自训练方式并未带来实质改善；且自训练反而在已解决的问题上造成了高于测量下限的退化率。因此，该研究强调，可靠的模型改进评估必须为每项统计结果提供独立验证的零假设，且这些零假设应基于多臂实验中已有的基准复现数据构建，而非依赖于单次运行或不足的样本量。
链接: https://arxiv.org/abs/2608.20290
作者: Cheng Xu,Nan Yan,Liming Chen,M-Tahar Kechadi
机构: University College Dublin(都柏林大学); Georgia Institute of Technology(佐治亚理工学院); Dalian University of Technology(大连理工大学)
类目: Artificial Intelligence (cs.AI); Computation and Language (cs.CL)
备注: Code and evaluation artifacts are available at this https URL
Abstract:Whether a language model has improved itself is increasingly judged not by mean accuracy but by which individual problems it gains and loses. Tracking these transitions means differencing two noisy estimates, leaving them vulnerable to measurement artifacts. Auditing three rounds of rank- 32 LoRA self-training on Qwen3-8B against a frozen control pushed through the identical pipeline, we identify seven measurement failures, each of which inverts a reported finding when its control is absent. Several are standard practice. A ledger built on a single greedy decode manufactures capability changes on an untrained model, largely an artifact of inference batching; the expansion statistic separating acquisition from sharpening assigns that same model a rate of 0.280 . The natural threshold repair does not survive replication: estimated across the frozen comparisons such a design already contains, its null stays non-zero. We replace it with a per-problem exact test against a pooled baseline under false-discovery-rate control, which detects nothing on any held-out replicate and is unchanged under the multiple-testing rule, error rate and pool size. Applied to a ladder of arms matched in stream, volume and evaluation, the audit finds that external distillation improves problems the base model rarely reaches while three forms of self-training do not; a regression rejects this asymmetry as a by-product of distillation’s larger overall gain ( p 10^-8 ). On the far smaller set of problems the base model never reaches, the evidence is inconclusive, while self-training corrupts problems solved at baseline at rates well above the measured floor. Transition-level auditing therefore requires a separately measured null for every statistic it reports: nulls that cost no new experiments, built from baseline replicates a multi-arm study already owns, though not from as few as most possess.
[NLP-6] Inject Align Recover: Staged Post-Training for Retrieval-Free Document Knowledge Internalization
【速读】： 该论文旨在解决大语言模型在推理时无法访问源文档的情况下，对限定文档集合进行问答的难题，核心问题在于如何实现文档知识内化（document knowledge internalization）——即在不依赖检索的前提下，将固定语料库的知识转化为可被模型参数存储并使用的知识。其解决方案的关键在于提出一种三阶段后训练框架IAR（Inject, Align, and Recover），通过分离式策略实现高效的知识注入与能力保持：首先，在“Inject”阶段，将源文档转换为连续生成、重写及指令条件下的重构目标，以结构化方式注入文档知识；其次，在“Align”阶段，利用仅含答案的问答监督信号对模型进行微调，使其具备精准的问答行为对齐能力；最后，在“Recover”阶段，将领域适配后的模型与基础指令模型融合，恢复其在通用任务上的综合表现。该方法在多个主流模型家族（Llama、Phi、Qwen、SmolLM）与数据集（CC、CCI）上均显著优于传统微调方法，尤其在7/8的设置中超越基准方法，平均提升领域问答准确率3.6个百分点，通用能力（IFEval、MMLU、MSBench）平均提升12.1个百分点，展现出在领域内化与通用能力之间取得平衡的卓越性能。
链接: https://arxiv.org/abs/2608.20281
作者: Qian Kou,Xiaofeng Shi,Xiaosong Qiu,Hua Zhou
机构: BAI Institute of Artificial Intelligence (北京智谱人工智能研究院)
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
备注: 21 pages, 4 figures. Includes Supplementary Material Sections A–G. Qian Kou and Xiaofeng Shi contributed equally and are co-corresponding authors. Hua Zhou is the project leader
Abstract:Large language models often fail to answer questions about a bounded document collection when the source documents are not retrieved at inference time. We study this setting as document knowledge internalization: converting a fixed corpus into usable parametric knowledge for retrieval-free question answering. We propose IAR (Inject, Align, and Recover), a three-stage post-training framework that separates structured document knowledge injection, QA behavior alignment, and general ability recovery. Unlike conventional continued pretraining, Inject converts source documents into continuation, rewrite, and instruction-conditioned reconstruction objectives. Align then adapts the injected model with answer-only QA supervision, while Recover merges the domain-adapted model with the base instruction model to recover general capabilities. Across Common Corpus (CC) and CCI, and across Llama, Phi, Qwen, and SmolLM model families, IAR improves the domain-primary domain-general frontier for retrieval-free document internalization. In the main comparison, IAR improves over Vanilla SFT on all four reported metrics in 7 of 8 dataset-model settings, with average gains of 3.6 percentage points in domain QA accuracy and 12.1 percentage points in mean general performance across IFEval, MMLU, and MSBench. Extended CC baselines show that LoRA and FAPM can win individual general metrics, but among methods that also reach leading or near-leading domain internalization, IAR retains one of the strongest general profiles.
[NLP-7] Break It Down Pass It On: Cross-Task Skill Transfer in LLM Agents
【速读】： 该论文旨在解决大语言模型（Large Language Model, LLM）代理在任务执行中通过经验归纳技能后，这些技能在跨任务迁移过程中存在不可靠甚至有害的问题。当前关键挑战在于：如何确保所诱导的技能能够稳定、有效且安全地在不同任务间迁移。论文的核心解决方案在于系统性地分析两种关键因素对技能迁移能力的影响：一是技能归纳的粒度（任务级与子任务级），二是技能表达形式（文本与代码）。研究发现，子任务级技能相较于任务级技能显著提升整体性能，而文本格式技能的迁移效果优于代码格式。为深入理解这一现象，作者提出两个互补属性——“特异性”（衡量技能与实际任务的匹配程度）与“抽象性”（衡量技能在各类任务间的适用广度），并构建了“技能效用评分”（skill utility score）以综合评估其迁移潜力。该评分仅依赖技能本身及其任务描述，无需实际执行过程，因此可作为部署新任务前对技能记忆质量的高效诊断工具。实证结果表明，该评分与任务成功高度相关，且子任务级和文本形式的技能得分更高，揭示了优化技能归纳策略的关键路径。
链接: https://arxiv.org/abs/2608.20274
作者: Yiyang Feng,Biddut Sarker Bijoy,Niranjan Balasubramanian,Jiawei Zhou
机构: Stony Brook University (石溪大学)
类目: Artificial Intelligence (cs.AI); Computation and Language (cs.CL)
备注: 34 pages, 28 figures, 7 tables
Abstract:Large language model (LLM) agents can induce skills from completed tasks and reuse them later to grow more capable with experience. In practice, induced skills may transfer unreliably and can even harm the agent that retrieves them. When agent-induced skills transfer reliably across tasks remains an open question. We conduct a comprehensive and controlled study of how the way skills are induced shapes their transfer across tasks. Specifically, we compare task-level with subtask-level skill induction and text with code skill formats, the two axes along which existing methods differ. Task-level skills mostly reduce the agent’s performance below its no-memory baseline while subtask-level skills raise it above on average, and text skills transfer better than code skills. To further understand our findings, we examine two complementary properties of the induced skills: specificity, which measures how closely a skill matches real tasks, and abstractness, which measures how evenly its relevance spreads across tasks. Neither property alone predicts task success, but their combined effect does, which we propose as a skill utility score. The score correlates consistently with task success when skills are transferred, and subtask-level and text skills score higher. Computing skill utility only needs the skills and task descriptions but not any task execution, so our score serves as a practical diagnostic of a skill memory before any new task runs.
[NLP-8] ContractScrub: A benchmark for final review of legal contracts ICML
【速读】： 该论文旨在解决法律领域中合同审查（contract scrubbing）任务自动化评估缺失的问题，该任务作为典型高文本处理强度的重复性工作，具有高度自动化潜力，但此前缺乏针对大语言模型（LLM）在该场景下表现的系统性评测。其解决方案的关键在于提出首个专门针对合同审查能力的基准测试——ContractScrub，该基准由经验丰富的律师手工构建，涵盖定义术语误用、引用错误、语言不一致等多样化的错误类型。实验表明，尽管前沿模型在通用基准上表现良好，但在ContractScrub上整体性能有限，仅有一款模型达到0.75的宏平均召回率（macro average recall），凸显当前模型在真实法律场景下的实际局限性，强调了构建领域特定、任务聚焦的评估基准对于准确衡量生成式AI（Generative AI）在复杂专业任务中真实效能的重要性。
链接: https://arxiv.org/abs/2608.20204
作者: Yejin Bang,Kirsty Fielding,Brandan Oliver,Brian Birke,Nabeel Seedat,Andrew M. Bean
机构: 未知
类目: Artificial Intelligence (cs.AI); Computation and Language (cs.CL)
备注: 10 pages, ICML AI4Law Workshop
Abstract:Legal work, with its heavy reliance on processing large amounts of text, is often considered one of the domains most exposed to the use of LLMs. Contract ``scrubbing,‘’ the final review of transactional agreements for errors and inconsistencies, is a particularly suitable task for automation, because it is routine, painstaking work requiring detailed attention to long documents. Scrubbing also seems to align naturally with the general capabilities expected of frontier LLMs around long-context reasoning, consistency checking, and named entity recognition (NER). Despite the economic value and potential for automation, no formal evaluations of LLMs performing contract scrubbing have been conducted. We introduce ContractScrub, the first benchmark designed to evaluate contract scrubbing capabilities, comprising contracts hand-crafted by experienced lawyers over diverse error categories such as misuse of defined terms, incorrect references, and inconsistent language. Frontier models perform surprisingly poorly with only one model reaching 0.75 macro average recall despite strong performance on seemingly related general benchmarks, demonstrating the practical limits of current models and the importance of narrowly targeted, domain-specific benchmarks for measuring real-world impact.
[NLP-9] MemTrapBench: Benchmarking Cognitive Traps in LLM Memory Use
【速读】： 该论文旨在解决现有大语言模型（Large Language Models, LLMs）记忆机制评估中忽略记忆对当前任务推理过程负面影响的问题。传统记忆基准主要关注信息的正确存储与提取，却未充分考察被检索的记忆如何扭曲模型的推理路径或改变其信念体系，进而导致任务性能下降。为此，论文识别出一类关键问题——记忆诱发的认知陷阱（Memory-induced Cognitive Traps），包括“推理固化”（Reasoning Fixation）和“信念扭曲”（Belief Distortion）。为系统化评估此类失效模式，作者构建了MemTrapBench基准测试框架。实验结果表明，所有被测记忆策略在该基准上均劣于无记忆设置，最强方法亦出现超过10%的性能下降。针对此问题，论文提出一种轻量级、高效的推理时干预方法AdaptiveMem，通过引导模型在推理过程中主动规避潜在记忆陷阱，在有效缓解认知陷阱的同时，保持甚至提升标准记忆基准上的表现，展现出良好的通用性与实用性。
链接: https://arxiv.org/abs/2608.20202
作者: Mengru Wang,Haozhe Luo,Zhenqian Xu,Zhixiang Cui,Haoming Xu,Qu Yang,Jizhan Fang,Junfeng Fang,Ningyu Zhang
机构: 未知
类目: Artificial Intelligence (cs.AI); Computation and Language (cs.CL); Computers and Society (cs.CY); Databases (cs.DB); Machine Learning (cs.LG)
备注: Work in progress
Abstract:Memory has become a key component of large language models, enabling them to retain information and learn from long-term interactions. However, existing memory benchmarks mainly evaluate whether information is correctly extracted, stored, and retrieved, while largely overlooking how retrieved memories reshape model reasoning and affect performance on the current task. We identify memory-induced cognitive traps: even faithfully recorded and semantically relevant memories can distort model reasoning or beliefs and degrade current task performance. To systematically evaluate these failure modes, we introduce MemTrapBench, which covers two forms of cognitive traps: Reasoning Fixation and Belief Distortion. Experiments across two model families and five representative memory frameworks show that MemTrapBench is challenging: all evaluated memory strategies underperform the no-memory setting, with even the strongest methods suffering drops of more than 10%. To mitigate these cognitive traps, we propose AdaptiveMem, a simple yet effective inference-time method that instructs LLMs to avoid memory traps. AdaptiveMem mitigates cognitive traps on MemTrapBench while preserving or improving performance on standard memory benchmarks across diverse memory frameworks.
[NLP-10] ask-CoEvolve: Efficient Harness Optimization via Adaptive Validation Task Selection
【速读】： 该论文旨在解决大语言模型（Large Language Model, LLM）智能体钩子（harness）优化过程中因固定全量验证集评估带来的高计算开销问题。现有方法在每次迭代中均对完整验证集进行评估，但随着钩子代码的演化，部分任务逐渐失去区分能力，仍被重复评估造成资源浪费。其解决方案的关键在于提出一种名为Task-CoEvolve的自适应验证任务选择方法，通过协同演化验证任务与钩子代码来提升效率。核心机制包括：一是基于历史评估结果采用方差加权采样策略，聚焦于候选钩子间分歧较大的任务（即位于智能体能力边界的任务），以最大化信息增益；二是通过引入采样概率校正的分数估计方法，从部分采样任务中推断出全集性能，确保不同迭代间评估结果的一致可比性。实验表明，Task-CoEvolve在在线文本分类和Terminal-Bench 2.1基准上显著优于固定子集基线方法，在达到与全集搜索相当的最终性能的同时，将评估次数减少了80%。
链接: https://arxiv.org/abs/2608.20169
作者: Atsuyuki Miyai,Kiyoharu Aizawa,Toshihiko Yamasaki
机构: The University of Tokyo (东京大学)
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
备注: Github: this https URL
Abstract:We present a novel approach to efficient LLM agent harness optimization through adaptive validation task selection. Harness optimization iteratively rewrites the harness code based on validation performance, enabling substantial performance gains without updating the underlying model weights. Existing approaches, however, evaluate a fixed validation set in full at every iteration, incurring substantial evaluation costs even on tasks that become less discriminative as the harness evolves. We propose \textbfTask-CoEvolve , which co-evolves the validation tasks with the harness by addressing two challenges: selecting informative tasks and estimating full-set performance from partial evaluations. Task-CoEvolve builds on the observation that tasks on which candidate harnesses disagree are more informative for distinguishing among them than tasks that are consistently solved or failed. It uses variance-weighted sampling based on past outcomes to focus evaluation on tasks near the agent’s capability frontier, with the sampling distribution adapting as the harness evolves. It then estimates full-set scores from the sampled tasks by accounting for their sampling probabilities, enabling consistent comparisons across iterations despite evaluating different subsets. Experiments on online text classification and Terminal-Bench 2.1 show that Task-CoEvolve consistently outperforms fixed-subset baselines and matches the final performance of full-set search while reducing the number of evaluations during optimization by 80%. Code will be released at this https URL.
[NLP-11] FormalTCS: Benchmarking End-to-End Frontier Formal Theoretical Computer Science Research of Large Language Models
【速读】： 该论文旨在解决当前大型语言模型（Large Language Models, LLMs）在理论计算机科学（Theoretical Computer Science, TCS）研究中评估基准不贴近真实科研场景的问题。现有基准未能充分反映前沿TCS研究的复杂性，如严格的定义依赖、假设条件及证明链条。为此，作者提出\ourbenchmark，一个由领域专家验证的基准，包含2025–2026年被STOC、FOCS、SODA和COLT录用论文中的175个实例，完整保留原始论文的定义、假设与证明依赖关系，并配有专家验证的Lean形式化陈述与证明。实验表明，当前主流LLMs在端到端TCS研究任务中仍表现欠佳，其中自动形式化（autoformalization）成为最显著的瓶颈——最佳模型在将自然语言命题转化为正式定理陈述上的表现仅为11.5分，远低于其在已提供形式化陈述下的证明能力（Pass@8为28.6）。基于该基准，研究进一步构建了自动化TCS研究框架，实现新命题的生成、形式化、筛选与证明。在64个生成命题中，仅有6个通过专家评审与证明验证，揭示出除形式化能力外，模型缺乏足够的“研究品味”（research taste），即对重要性、新颖性和可证性的判断能力，仍是实现自主TCS研究的主要障碍。
链接: https://arxiv.org/abs/2608.20153
作者: Dingzirui Wang,Xuanliang Zhang,Keyan Xu,Qingfu Zhu,Wanxiang Che
机构: Harbin Institute of Technology (哈尔滨工业大学)
类目: Computation and Language (cs.CL)
备注:
Abstract:Large language models (LLMs) have shown growing potential for automated theoretical computer science (TCS) research, yet existing benchmarks remain far from realistic research settings. We introduce \ourbenchmark, an expert-validated benchmark for evaluating LLMs on frontier, end-to-end TCS research. \ourbenchmark contains 175 instances drawn from papers accepted to STOC, FOCS, SODA, and COLT in 2025-2026, preserving paper-specific definitions, assumptions, and proof dependencies, with expert-verified Lean formalizations and proofs. Evaluations of leading LLMs reveal that current models remain far from reliably completing the full research pipeline. In particular, autoformalization is the sharpest bottleneck: the best model achieves only 11.5 on translating natural-language claims into formal theorem statements, compared with 28.6 Pass@8 when proving human-provided formal statements. Building on \ourbenchmark, we further develop an automated TCS research framework that generates, formalizes, filters, and proves new claims. Of 64 generated claims, only 6 ultimately pass expert evaluation and proof verification, indicating that beyond formalization, limited research taste remains another major barrier to autonomous TCS research.
[NLP-12] When Text and Numbers Disagree: Evidence Arbitration in Large Language Models
【速读】： 该论文旨在解决大语言模型（Large Language Models, LLMs）在面对多源异构证据（如文本摘要、数值观测与外部工具输出）存在冲突时的决策仲裁问题。其核心挑战在于，当不同模态的信息支持相反结论时，模型如何权衡并做出最终判断。解决方案的关键在于构建一个受控的合成基准测试（synthetic benchmark），通过隐含的风险轨迹同时生成时间序列数值数据和自然语言摘要，从而精确控制各证据源与真实标签的一致性，实现对模态类型、时间新近性、来源可靠性及证据溯源等变量的独立操纵。实验结果表明，当前LLMs的仲裁行为具有系统性而非随机性：模型表现出明显的文本与数值偏好差异，更倾向于遵循时间新近性而非明确的可靠性线索，并可能过度依赖外部预测结果，即使其与直接上下文证据相矛盾。这揭示了现有大语言模型在整合异构证据时主要依赖启发式策略，暴露出工具增强型决策系统中的关键失效模式。
链接: https://arxiv.org/abs/2608.20116
作者: Mattia Carletti,Edward Phillips,Fredrik K. Gustafsson,Patitapaban Palo,Lei Clifton,Danielle Belgrave,Xiao Gu,David A. Clifton
机构: Nuffield Department of Primary Care Health Sciences, University of Oxford, Oxford, UK; GlaxoSmithKline, London, UK; Oxford Suzhou Centre for Advanced Research, University of Oxford, Suzhou, Jiangsu, China; Department of Engineering Science, University of Oxford, Oxford, UK
类目: Computation and Language (cs.CL)
备注:
Abstract:Large language models (LLMs) are increasingly used in settings where textual summaries, numerical observations, and external tool outputs may provide conflicting evidence. We study how LLMs arbitrate between such sources when they support opposing decisions. To do so, we introduce a controlled synthetic benchmark in which latent risk trajectories generate both numerical time series and natural language summaries, allowing us to construct conflicts where exactly one evidence source is aligned with the ground-truth label. This design lets us independently manipulate modality, temporal recency, source reliability, and evidence provenance. Across open-weight instruction-tuned models, we find that arbitration behaviour is systematic rather than random: models exhibit distinct text-versus-number preferences, follow temporal recency more consistently than explicit reliability cues, and can over-rely on external forecasts even when they conflict with direct contextual evidence. These results suggest that current LLMs often rely on heuristic arbitration strategies when integrating heterogeneous evidence, highlighting a failure mode for tool-augmented decision systems.
[NLP-13] OenoBench: A Wine-Domain Benchmark for Knowledge-Grounded Evaluation of Large Language Models
【速读】： 该论文旨在解决生成式 AI 在专业领域知识问答中存在事实可信度不足、评估体系不透明以及模型性能评估缺乏系统性基准的问题。其核心挑战在于如何构建一个高可信度、可追溯且多层级的垂直领域知识评测基准，以真实反映大语言模型（LLM）在葡萄酒专业知识领域的理解与推理能力。解决方案的关键在于提出并实现一个基于验证事实源的 LLM 驱动型流水线：所有知识均源自政府注册机构（如 INAO、TTB、OIV）、同行评审期刊及维基百科/维基数据等权威来源，并通过 35 个经过溯源验证的爬虫提取原子级事实；每个问题均通过五种生成策略和五类生成器组合生成，且每条陈述均可追溯至原始链接；最终采用九代理审计机制，基于人类黄金标准进行 Cohen’s κ 校准，确保评估结果的可靠性。该方法不仅保障了知识来源的真实性，还实现了对模型性能的精细化分析，揭示了不同模型在推理模式、自偏好行为及闭卷可解项上的表现差异，为后续研究提供了可复现、可验证的基准框架。
链接: https://arxiv.org/abs/2608.20106
作者: Nikita Khudov
机构: StrategAI
类目: Computation and Language (cs.CL)
备注:
Abstract:We introduce OenoBench, a wine-domain knowledge benchmark of 3,266 multiple-choice questions across six pillars (regions, grape varieties, viticulture, winemaking, producers, business) and four difficulty tiers. The corpus is built from 38,104 atomic, source-anchored facts extracted by 35 provenance-verified scrapers from government registries (INAO, TTB, OIV), peer-reviewed journals, and Wikipedia/Wikidata. Our methodological contribution is an LLM-driven pipeline in which language models reformat verified facts and audit the result, but never serve as the source of truth: every claim traces to a URL, every question is generated by one of five strategies across five generator families, and every question is scored by a nine-agent audit calibrated against a human gold sheet via Cohen’s \kappa . Evaluating sixteen frontier configurations, we find: (i) overall accuracy spans 53%-84%, led by o3 at 83.6%; (ii) reasoning-mode lift concentrates in DeepSeek R1 (+6.8pp) and is absent in Claude Opus and Gemini Pro; (iii) Anthropic shows +9pp self preference on its own questions while Google shows -8pp inverse preference; (iv) frontier open-weight models share the cost-vs-accuracy Pareto frontier with proprietary reasoning models; and (v) every config gains around 33pp on closed-book solvable items, revealing a parametric-recall ceiling that only the contextual slice avoids. We release corpus, audit findings, human-review app, and construction code under CC-BY-SA-4.0.
[NLP-14] SABET-QA: Temporal Knowledge Graph Question Answering
【速读】： 该论文旨在解决时间知识图谱问答（Temporal Knowledge Graph Question Answering, TKGQA）中多步推理难题，尤其是现有基于嵌入的方法在处理复杂时序查询时因单次遍历推理机制导致的性能瓶颈。其核心解决方案是提出SABET-QA框架，通过双向实体-时间评分机制实现跨多跳的推理状态迭代优化，并引入槽位感知的上下文建模模块以对齐问题语义与时间知识图谱嵌入表示。该框架采用可微分的工作记忆机制支持渐进式假设精炼，同时利用辅助的时间边界信息作为粗粒度监督信号，提升模型对时间约束的敏感性。实验在CronQuestions、Complex-CronQuestions、MultiTQ和TimeQuestions数据集上验证了该方法在复杂多步时序查询上的显著优势，优于多个强基准模型。
链接: https://arxiv.org/abs/2608.20083
作者: Brahim Touayouch,Mirette Moawad,Dmitry Akulov
机构: ENS Paris-Saclay(巴黎萨克雷国立高等理工学院); École Polytechnique(巴黎综合理工学院); QuickSort Research(快速排序研究)
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
备注:
Abstract:Question Answering over Temporal Knowledge Graphs (TKGQA) requires reasoning over time-sensitive facts, yet existing embedding-based methods struggle with multi-step queries due to single-pass reasoning pipelines. We propose SABET-QA, a framework that iteratively refines reasoning states across multiple hops via a bidirectional entity-temporal scoring mechanism and a slot-aware contextualization module that aligns question semantics with temporal KG embeddings. A differentiable working memory enables progressive hypothesis refinement, while auxiliary temporal boundaries serve as coarse supervision when available. Experiments on CronQuestions, Complex-CronQuestions, MultiTQ, and TimeQuestions demonstrate consistent improvements over strong baselines, particularly on complex multi-step temporal queries.
[NLP-15] Lets Scale Step by Step: Compute-Efficient Hyperparameter Transfer for Large-Scale Mixture-of-Experts
【速读】： 该论文旨在解决大规模混合专家（Mixture-of-Experts, MoE）模型在极端规模下的超参数优化难题，尤其是学习率（learning rate）的调优问题。由于在模型规模和训练词元预算（token budget）均达到万亿级别时，传统的超参数搜索方法因计算成本过高而不可行。其解决方案的关键在于提出一种计算高效的两阶段超参数迁移框架：首先，通过引入多头潜在注意力（Multi-head Latent Attention, MLA）与Muon优化器，构建最大更新参数化（Maximal Update Parameterization, μP），证明了最优学习率在不同宽度缩放（width-scaled）的MoE模型间具有稳定的可迁移性；其次，进一步将该迁移能力扩展至训练词元维度，建立预测性缩放定律（scaling law），利用小规模代理模型在有限预算下的最优学习率数据，通过线性回归实现对超大规模训练场景（如10万亿词元）下理想学习率的高精度外推（R²=0.95）。该方法表明，仅通过小规模代理模型的预训练即可准确预测大规模MoE模型的最优训练配置，显著降低全规模模型的调参成本。研究将其应用于从零开始预训练一个1550亿总参数、170亿活跃参数的基础模型，验证了该方法在实际应用中的有效性与稳定性。
链接: https://arxiv.org/abs/2608.20061
作者: Nayeon Kim,Hojin Lee,Yunju Bak,Jaesun Park,Boseop Kim
机构: Kakao Corp.(Kakao公司); Upstage AI
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Computation and Language (cs.CL)
备注: COLM 2026
Abstract:Mixture-of-Experts (MoE) architectures significantly expand model capacity without a proportional increase in computational cost. However, optimizing their hyperparameters—particularly the learning rate—at extreme scales of both model size and token budget via sweeping remains computationally prohibitive. In this paper, we propose a compute-efficient, two-step hyperparameter transfer framework that estimates optimal learning rates for training large MoE models by transferring them across scaling model widths, and subsequently extrapolating to trillion-token horizons. First, we formulate a Maximal Update Parameterization ( \mu P) adaptation for MoE architectures utilizing Multi-head Latent Attention (MLA) and the Muon optimizer, demonstrating that optimal learning rates transfer consistently across width-scaled models. Second, we extend this transferability along the token dimension by establishing a predictive scaling law. By applying linear regression to the optimal values derived from small proxy models on limited budgets, we successfully extrapolate the ideal learning rate to massive training horizons (e.g., 10 trillion tokens) with high fidelity ( R^2=0.95 ). Consequently, this indicates that proxy training on small models is sufficient to determine the optimal learning rate for the extensive training of large-scale MoEs. We apply the proposed methodology to pretrain our foundation model (155B total, 17B active parameters) from scratch, and the stable training and evaluation results validate that optimal configurations for full-scale target models can be accurately predicted with minimal ablation costs.
[NLP-16] Auditing Cross-Lingual Fairness in Language Model Watermarking
【速读】： 该论文旨在解决现有大语言模型输出水印方案评估体系在多语言场景下存在的局限性问题。当前的评估方法主要局限于英语文本，依赖单一检测阈值和有限的质量度量指标，导致在跨语言部署中隐含的评估设计选择（如阈值设定、质量评估范式）对结果的影响被忽视，进而可能误导对水印方案公平性和有效性的判断。其解决方案的关键在于提出一个系统化的多语言水印评估框架，包含四个核心组件：基于具体部署场景的经验校准检测阈值、与阈值无关的辅助度量以区分校准失败与检测失败、三种独立的质量评估范式（分布一致性、成对语义相似性、参考困惑度），以及基于类型学家族划分的广义熵分解方法，用于量化跨语言差异。该框架通过在六种水印方案、三个开源生成器、十一种涵盖四种书写系统和八种类型学家族的语言及基础与指令微调两种模式下的实证分析，揭示了单语言、单范式评估无法捕捉的失效模式。研究发现，检测性能与质量表现的跨语言差异主要存在于类型学家族之间，表明水印技术中的跨语言公平性差距本质上源于语言本身的结构性特征，而非特定语言的偶然偏差。
链接: https://arxiv.org/abs/2608.20047
作者: Alexander Nemecek,Osama Zafar,Debargha Ganguly,Vikash Singh,Vipin Chaudhary,Erman Ayday
机构: Case Western Reserve University (凯斯西储大学)
类目: Computation and Language (cs.CL); Cryptography and Security (cs.CR); Machine Learning (cs.LG)
备注: 24 pages
Abstract:Watermarking schemes for large language model output are evaluated almost exclusively on English text using each scheme’s detection threshold and a narrow set of quality measurements. Multilingual deployment exposes evaluation-design choices that are inconsequential on English but determine conclusions cross-lingually. We propose an evaluation framework with four components: detection thresholds calibrated empirically per deployment context, a threshold-independent companion measurement that distinguishes calibration failures from detection failures, three disjoint quality measurement paradigms (distributional, paired-semantic, and reference-perplexity), and a generalized-entropy decomposition of cross-language disparity over a typological family partition. Applied to six watermarking schemes, three open-weight generators, eleven languages spanning four scripts and eight typological families, and both base and instruction-tuned regimes, the framework reveals failure modes that single-language single-paradigm evaluation cannot surface. Across detection and quality, observed disparity is predominantly between-family on the typological partition, indicating that cross-lingual fairness gaps in watermarking are structural to language properties rather than idiosyncratic to particular languages.
[NLP-17] HealMed: Multilingual Evaluation of Large Language Models in Medicine
【速读】： 该论文旨在解决多语言医学领域大语言模型（Large Language Models, LLMs）评估缺乏统一、权威基准的问题。现有评估体系在多语言场景下存在覆盖不足、翻译质量参差不齐及评估结果不可靠等缺陷，尤其在低资源语言中表现尤为显著。为此，论文提出HealMed——一个由23名来自九个国家和地区的医学专家历时两年审校的多语言医学评估基准，涵盖九种语言、1000个样本，覆盖选择题（MCQA）、自然语言推理（NLI）和开放问答（open-ended QA）三类任务形式。其解决方案的关键在于：通过双语专家对每项翻译进行严格评审与修订，确保跨语言数据的语义一致性和临床准确性；同时揭示了模型性能在低资源语言中的显著下降趋势，并发现仅具备医学专业化训练不足以保证多语言鲁棒性，而翻译质量对跨语言评估结果具有决定性影响，凸显高质量人工校验在构建可信多语言评估基准中的核心作用。
链接: https://arxiv.org/abs/2608.19981
作者: Yingjian Chen,Fan Gao,Sherry T. Tong,Haoyu Zhang,Aosong Feng,Kevin W. Jin,Xing Wu,Jinghui Lu,Abdul Samad,Akbar Faruqi,Cesar Caraballo,Cibele Brandão,Dhruva(Drew)Gupta,Eunji Jeon,Gabriel Madera-Santiago,Geon Lee,Hugo Toshio Itikawa,Insook Cho,Isabelli Martins,Isarar Siddique,Israr Ahmed,Jihyo Kwak,Kanyakorn Veerakanjana,Luis Guilherme Cardoso,Minjin Kim,Piyalitt Ittichaiwong,Renee Dua,Santiago Gudiño-Rosales,Xiujie Chen,Zeo Lapalus,Zixin Xu,Michihiro Yasunaga,Rex Ying,Heuiseok Lim,Jaewoo Kang,Chanjun Park,Hang Jiang,Ethan Goh,Hyunjae Kim,Edison Marrese-Taylor,Yusuke Iwasawa,Yutaka Matsuo,Qingyu Chen,Irene Li
机构: HealMed
类目: Computation and Language (cs.CL)
备注:
Abstract:We present HealMed, an expert-reviewed benchmark for multilingual evaluation of large language models in medicine. HealMed contains 1,000 examples in each of nine languages, drawn from nine datasets and covering three task formats: MCQA, NLI and open-ended QA. The benchmark was developed over two years by 23 physicians and medical experts based across nine countries and regions. Each translation was evaluated and revised by two experts fluent in English and the corresponding target language. On HealMed, performance declined most in low-resource languages, although the size of the gap varied markedly across languages and models. The strongest proprietary models were the most stable across languages, whereas many open-source and medically specialized models showed larger and less consistent gaps. Medical specialization alone did not ensure multilingual robustness. Furthermore, expert revision could either raise or lower measured performance, indicating that translation quality materially affects cross-language evaluation results.
[NLP-18] Robust Incomplete Multimodal Sentiment Analysis via Iterative Proxy Correction
【速读】： 该论文旨在解决真实场景下多模态情感分析（Multimodal Sentiment Analysis, MSA）中因模态缺失或数据损坏导致的跨模态互补性减弱及误导信息引入问题。现有基于代理（proxy）的方法通常依赖一次性构建代理来补全受损的语言信息，但初始生成的代理可能粗糙且不可靠，过早注入多模态推理过程会传播初始误差，影响情感预测性能。其解决方案的关键在于提出一种迭代式代理校正框架：通过非语言模态构建以语言为导向的代理，并在多模态上下文中利用门控残差校正机制逐步优化代理表示；同时，根据估计的语言可靠性得分，自适应地融合校正后的代理与可观测语言表征，实现代理补偿与可信语言证据之间的动态平衡。此外，引入分阶段潜在校正目标，以完整语言表示作为训练阶段的语义锚点，稳定代理优化轨迹。实验结果表明，该框架在MOSI、MOSEI和SIMS数据集上于多种模态缺失设置下均显著优于现有基线，展现出对不完整输入的鲁棒情感预测能力。
链接: https://arxiv.org/abs/2608.19971
作者: Zhifa Geng,Subin Huang,Hao Guo,Junjie Chen,Sanmin Liu,Chao Kong
机构: Anhui Polytechnic University (安徽工程大学)
类目: Computation and Language (cs.CL)
备注: Accepted to SEKE 2026. 6 pages, 4 figures
Abstract:Multimodal sentiment analysis aims to infer affective states by integrating language, visual, and acoustic cues. However, real-world multimodal inputs are often incomplete or corrupted, which can weaken cross-modal complementarity and introduce misleading information into downstream fusion. Existing proxy-based methods for incomplete MSA commonly rely on one-shot proxy construction to compensate for degraded language information, but the generated proxy may be coarse or unreliable at initialization. Prematurely injecting such a proxy into multimodal reasoning can propagate initial errors and compromise sentiment prediction. To address this limitation, we propose an iterative proxy correction framework for robust incomplete MSA. Our method constructs a language-oriented proxy from non-language modalities and progressively refines it under multimodal context through gated residual correction. The corrected proxy is then adaptively fused with the observed language representation according to an estimated language reliability score, allowing the model to balance proxy-based compensation and trustworthy linguistic evidence. In addition, we introduce a stage-wise latent correction objective that uses the complete language representation as a training-time semantic anchor to stabilize the proxy refinement trajectory. Extensive experiments on MOSI, MOSEI, and SIMS under diverse missing-modality settings demonstrate that the proposed framework consistently outperforms competitive baselines and achieves robust sentiment prediction under incomplete inputs.
[NLP-19] Natural Language Code Retrieval for 1C:Enterprise: An Open Benchmark and Efficient Bi-Encoder
【速读】： 该论文旨在解决1C:Enterprise生态系统中自然语言代码检索（Natural Language Code Retrieval）缺乏公开数据集与专用模型的问题，其核心挑战在于该系统采用俄语语法并包含高度领域特定的术语，导致现有通用方法难以适用。解决方案的关键在于构建一个完整的端到端管道：首先提出一个包含3,413个真实世界、已去除个人身份信息（PII）的查询-代码配对的开源基准数据集；其次设计可复现的评估框架；最后开发一种专用于该领域的双编码器（bi-encoder）模型。为应对标注数据稀缺问题，研究者利用Google Gemma-4-26B-A4B-it模型从公共代码仓库生成784,057组合成三元组进行微调，并结合马特里什卡表征学习（Matryoshka Representation Learning, MRL）与隐私感知分词器（privacy-aware tokenizer），有效提升模型在低资源场景下的泛化能力。实验结果表明，所提模型在平衡子集宏平均nDCG@10上达到0.5992，显著优于基线架构（0.4932）和Google EmbeddingGemma-300M（0.5404）。进一步分析显示，即使剔除所有经严格精确匹配/13-gram重叠审计标记的训练-测试重叠样本，模型性能仍保持在0.6011（平衡宏平均），证明性能优势并非由数据泄露所致。此外，通过将MRL表征维度压缩至256维，可在保留99.9%检索质量的同时，使密集索引存储与精确相似性计算量降低三个数量级，显著优化部署效率。
链接: https://arxiv.org/abs/2608.19957
作者: Konstantin Chesnokov,Chingiz Mingazov
机构: Independent Researcher(独立研究员); Independent Researcher(独立研究员)
类目: Computation and Language (cs.CL)
备注:
Abstract:Natural language code retrieval is a rapidly evolving task in computer science. However, the 1C:Enterprise ecosystem combines Russian syntax with highly domain-specific terminology, for which open datasets and specialized models have been virtually non-existent. We present a comprehensive pipeline for 1C code retrieval: an open benchmark of 3,413 real-world, PII-scrubbed query-code pairs, a reproducible evaluation harness, and a specialized bi-encoder. To overcome scarce labeled data, we fine-tune on 784,057 synthetic triplets generated by google/gemma-4-26B-A4B-it from public code repositories, using Matryoshka Representation Learning (MRL) and a privacy-aware tokenizer. Because the benchmark subsets differ in size, we report balanced-subset macro, query-weighted micro, and forum-only results. Our model reaches 0.5992 balanced macro nDCG@10, 0.5044 micro, and 0.4617 on forum, versus 0.4932 macro for the baseline architecture and 0.5404 for google/embeddinggemma-300m. Removing every benchmark example flagged by the conservative exact/13-gram overlap audit leaves 0.6011 balanced macro (0.5010 micro), indicating that detected train-benchmark overlap does not explain the headline result. MRL truncation to 256 dimensions preserves 99.9% of retrieval quality while reducing dense-index storage and exact similarity arithmetic by a factor of three.
[NLP-20] Dynamic Gated Cross-Modal Fusion with Sarcastic-aware Contrastive Regularization for Multimodal Sarcasm Detection
【速读】： 该论文旨在解决多模态讽刺检测（Multimodal Sarcasm Detection, MSD）中因模态间贡献动态变化及表面语义一致性掩盖深层矛盾意图而导致的检测难题。现有方法通常采用固定融合策略，将讽刺简单视为跨模态不一致，难以捕捉细微的讽刺线索与实例特定的模态交互。其核心解决方案是提出一种集成动态门控跨模态融合（Dynamic Gated Cross-Modal Fusion）与讽刺感知对比正则化（Sarcastic-aware Contrastive Regularization, SaCR）的新框架。关键在于：通过双向门控交互模块实现文本与视觉特征的实例级自适应过滤与贡献校准，并引入动态融合门以平衡模态重要性；同时，SaCR作为标签感知的对比正则化目标，促使非讽刺样本保持语义一致性，而抑制讽刺样本中的误导性一致表征，从而更精准地揭示讽刺意图。整个框架采用多目标学习策略端到端训练，联合优化多模态分类与辅助单模态监督任务，在MMSD和MMSD2.0数据集上均显著优于现有基线方法。
链接: https://arxiv.org/abs/2608.19942
作者: Hao Guo,Subin Huang,Junjie Chen,Zhifa Geng,Sanmin Liu,Chao Kong
机构: Anhui Polytechnic University (安徽工程大学)
类目: Computation and Language (cs.CL)
备注: Accepted to SEKE 2026. 6 pages, 3 figures
Abstract:Multimodal sarcasm detection aims to identify sarcastic intent from multimodal content, where inconsistencies between literal meaning and contextual cues often signal irony. This task has attracted increasing research attention. However, accurate detection remains challenging due to instance-dependent modality contributions and misleading semantic consistency, where surface-level alignment masks underlying contradictory intent. Existing methods often rely on fixed fusion strategies and treat sarcasm as generic cross-modal mismatch, limiting their ability to capture subtle sarcasm cues and instance-specific modality interactions. To address these challenges, we propose a novel MSD framework that integrates Dynamic Gated Cross-Modal Fusion with Sarcastic-aware Contrastive Regularization (SaCR). Specifically, a bidirectional gated interaction module performs cross-modal feature filtering and adaptively calibrates textual and visual contributions at the instance level. A dynamic fusion gate further balances modality importance to generate more robust multimodal representations. Furthermore, SaCR is introduced as a label-aware contrastive regularization objective that encourages semantic consistency for non-sarcastic samples while suppressing misleading consistency in sarcastic cases. The proposed framework is trained end-to-end with a multi-objective learning strategy that jointly optimizes multimodal classification and auxiliary unimodal supervision. Extensive experiments on MMSD and MMSD2.0 demonstrate that the proposed method consistently outperforms strong baselines.
[NLP-21] Learning how to Forget: Fine-tuning for Long-Context Sparse Attention
【速读】： 该论文旨在解决在保持较低硬件资源消耗的前提下，实现基于稀疏注意力（Sparse Attention）的长上下文推理中关键值（Key-Value, KV）缓存选择与压缩所面临的模型训练难题。传统方法在使用稀疏注意力策略时，往往难以有效适配模型自身特性，导致性能下降。本文提出了一种新的微调方法，其核心在于使模型能够与任意KV缓存策略协同优化（co-adapt），从而在不依赖高成本硬件（如多卡并行）的情况下，仅需单块NVIDIA A100 GPU（40 GB显存）即可完成高效训练，并显著提升模型在长序列任务上的表现，通常优于采用精确注意力（Exact Attention，即序列并行）训练的模型。解决方案的关键在于设计了一种兼容性强、可与不同缓存策略联合优化的微调机制，同时结合针对H2O稀疏注意力策略的高效实现，通过专用缩放点积注意力核（scaled dot product attention kernel）优化计算效率。此外，作者开源了KeysAndValues库，为长上下文推理与微调提供了高性能、易用的代码支持。
链接: https://arxiv.org/abs/2608.19920
作者: Matthias Seeger,Zeyu Zhang,Vihang Patil,Konstantinos Benidis,Sebastian Schelter
机构: Amazon Web Services(亚马逊网络服务); University of Amsterdam(阿姆斯特丹大学); Amazon(亚马逊); Technical University Berlin(柏林工业大学)
类目: Computation and Language (cs.CL)
备注: 39 pages, no figures
Abstract:A lot of prior work addressed key-value (KV) cache selection and compression by sparse attention to enable long-context inference for transformer language models without excessive hardware budgets. We provide a new method for fine-tuning models with sparse attention. It works for any KV cache policy, runs on a moderate hardware budget (e.g., a single Nvidia A100 GPU with 40 GB RAM), and allows the model to co-adapt with the policy, often outperforming models trained with exact attention (sequence parallelism). We also provide an efficient implementation of H2O sparse attention (the leading policy in our experiments) with dedicated scaled dot product attention kernel support. KeysAndValues (this https URL), a new open source library for long-context inference and fine-tuning, provides easy-to-use and performant code for all methods discussed here.
[NLP-22] Interrupting the Loop: Periodic Subject Changes Raise Judged Surprise and Connection in Base Language Models
【速读】： 该论文旨在探究基础语言模型（Large Language Model, LLM）在无任务设定下生成内容的创新性来源，以及模型对长文本流的实际感知能力。其核心问题在于：生成内容中的“新颖性”究竟从何而来？模型在持续生成过程中是否能有效捕捉并维持上下文连贯性？为解决这一问题，研究提出了一种受认知启发的生成循环实验框架，在三种基础模型上测试了24种条件。关键发现是，生成新颖性的主要驱动力源自每数百个词注入一次的新主题（即“中断”），而这种中断通过抑制文本的字面重复（习惯化）来增强生成内容的意外性与连贯性。研究采用窗口级评估协议，以前提句为单位（n=10），通过可重复性检验对比人工评判者与第二组评判者，结果表明，相较于仅依赖习惯化的生成方式，引入中断可使主观惊喜度提升1.2至1.4分，连贯性提升0.8分。此外，研究发现：要求连续性的连贯性约束反而损害表现；单纯段落分隔对新文本无显著影响；重置上下文环境至少不劣于保留上下文；且在新前提下的预注册复制实验验证了核心对比的有效性。值得注意的是，有三项窗口评判者无法观测到的变化——如实验者注入句被误判为模型自生成、固定轮换注入句导致模型回放超出评判视野的内容、局部增益无法累积形成整体整合文档——揭示了当前评估机制的局限性。研究进一步指出，尽管引入显性注意力监控、闭环评判、跨中断记忆及评判门控的审查运行等机制均未带来实质提升，但在一个需验证器的在线装箱问题中，中断干预仍使有效且多样化的启发式策略数量增加三至四倍，而最优解质量未下降。因此，本研究并非揭示创造性的内在机制，而是提出了一种针对长文本生成的可控评估协议，并系统刻画了一种简单但有效的干预手段——即周期性中断——在激发多样性与新颖性方面的关键作用。
链接: https://arxiv.org/abs/2608.19893
作者: Roberto I. Ono Filho
机构: 独立研究员(Independent researcher)
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
备注: 48 pages including appendix; code, data pipeline and lab notebook at this https URL
Abstract:Where does the novelty a base language model produces with no task come from, and what can an LLM judge of a long stream actually see? We dismantle a cognitively inspired generation loop over 24 conditions on three base models. Most of its effect lives in one operation: a new subject injected every few hundred tokens (an interruption) into a stream whose literal repetition is damped (habituation). We judge windows of generated text only, with the premise as the unit (n=10) and a judge measured for repeatability, against a second judge family and against human readers. Under that protocol the interruption raises judged surprise by 1.2 to 1.4 points and connection by 0.8 over habituation alone. A connective that asks for continuity hurts; a bare paragraph break adds nothing detectable on fresh text; a reset context does at least as well as a kept one; and a pre-registered replication on new premises confirms the primary contrast. Three things the window judge could not see changed the first version of this study, and we think they are of general use. The judge scores the experimenter’s injected sentence as the model’s own. A fixed rotation of injected sentences makes the model replay its earlier segments from beyond the judge’s horizon, and the judge scores the replay as surprise and connection (65-80% of post-interruption windows at periods 150-300). And the local gains do not compose: no arm produces an integrated document. The salience monitor, the in-loop judge, memory across interruptions and a judge-gated Review run with a gate that opens add nothing. On a problem with a verifier (online bin packing), the interruption multiplies valid, distinct candidate heuristics three- to fourfold without raising the quality of the best. We report an evaluation protocol for long generation and a controlled characterization of a simple intervention, not a mechanism of creativity.
[NLP-23] EnvHarness: Awakening Static Worlds for Agent Learning
【速读】： 该论文旨在解决大语言模型（Large Language Model, LLM）智能体在静态、人工构建环境中的训练局限性问题。传统环境因缺乏对智能体弱点的感知且无法随其能力提升而动态演化，导致训练过程逐渐失效。现有环境生成方法虽试图缓解此问题，但通常依赖特定领域的工作流、昂贵或不可靠的验证机制，并仍生成静态环境。为此，本文提出可编程的环境调制框架——环境调制器（Environment Harness, EnvHarness），通过插件化组件层封装静态环境，在不修改底层逻辑的前提下重构其行为，实现跨领域的通用适配，并保持原有验证器的完整性。为实现自动化，进一步提出EnvRigger，将目标策略视为黑箱，通过分析其执行轨迹识别缺陷并自动合成针对性的EnvHarness组件，再通过新采样回放进行验证。在四个领域的五个基准测试中，EnvHarness显著优于原始环境及领域专用环境生成方法，在保留实例上最高提升9.0分，同时减少9.8%的执行步数。更重要的是，该框架为强化学习提供了更优的优化信号，支持策略与环境的持续、定向共进化。
链接: https://arxiv.org/abs/2608.19880
作者: Chengsong Huang,Zifeng Wang,Rujun Han,Jun Yan,Yanfei Chen,Zoey CuiZhu,Ke Jiang,Peng Xia,Han Yu,Yufan Zhuang,Yifei Ming,Jiaqi Pan,Bhavana Dalvi Mishra,Jiaxin Huang,Burak Gokturk,Tomas Pfister,Chen-Yu Lee
机构: 未知
类目: Artificial Intelligence (cs.AI); Computation and Language (cs.CL); Machine Learning (cs.LG)
备注:
Abstract:LLM agents learn by interacting with environments, yet these environments are hand-built and static: blind to an agent’s weaknesses, and quickly left behind as it improves. While recent environment generation methods attempt to address this, they require domain-specific pipelines, rely on expensive or unreliable verifiers, and still produce static environments. To alleviate the engineering burden of rebuilding environments from scratch, we propose Environment Harness (EnvHarness), a programmable layer of plug-in components that wraps a static environment to reshape its behavior without modifying the underlying logic. Operating through standard interfaces, EnvHarness applies across diverse domains while ensuring every reshaped environment retains its original verifier. To automate this process, we introduce EnvRigger, which treats the target policy as a black box, observing its execution trajectories to synthesize EnvHarness components targeting diagnosed flaws, and validating them via fresh rollouts. Across five benchmarks in four domains, EnvHarness outperforms both original environments and domain-specific environment generation pipelines, achieving up to a 9.0-point improvement on held-out instances with 9.8% fewer execution steps. Furthermore, EnvHarness provides a superior optimization signal for reinforcement learning, enabling continuous, targeted co-evolution of the policy and its environment.
[NLP-24] A knowledge-guided agent ic framework for mitigating patient-context ambiguity in health queries
【速读】： 该论文旨在解决医疗聊天机器人在面对患者提交的简短、信息不足的查询时，因缺乏患者特定上下文（如症状、诊断、用药史、过敏史或饮食限制等）而导致回答歧义的问题。此类查询虽语言清晰，但可能对应多个合理答案，直接由大语言模型响应易基于未经证实的假设生成不准确或不安全的回答。其解决方案的关键在于提出一种基于知识图谱的代理式（agentic）框架，该框架在原始语言模型之前介入，通过解析初始查询，利用任务特定的知识图谱构建一系列合理的假设，识别区分这些假设所必需的缺失患者上下文变量，并主动提出针对性的追问以获取关键信息。随后，将原始查询与新获取的上下文融合为澄清后的提示（clarified prompt），输入下游语言模型生成更精准的回答。实验结果表明，该框架在诊断检索和饮食安全性分类两项基准测试中显著优于直接回答或仅重述查询的方法，尤其在提升精确度（Top-1准确率）和召回率（Recall@5）方面表现突出，验证了其在缓解患者上下文模糊性方面的有效性。
链接: https://arxiv.org/abs/2608.19875
作者: Mahyar Abbasian,Saba A. Farahani,Arshia Ilaty,Hung Cao,Ramesh Jain,Amir M. Rahmani
机构: 未知
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
备注: 48 pages, 3 figures, 6 tables, journal
Abstract:Patients often submit short, underspecified queries to healthcare chatbots that lack the patient-specific information needed to determine an appropriate response. Although these queries may be linguistically clear, they can support multiple plausible answers depending on undisclosed factors such as symptoms, diagnoses, medications, allergies, or dietary restrictions. A language model answering such a query directly may therefore rely on unsupported assumptions about the patient. We introduce a knowledge-guided agentic framework for mitigating patient-context ambiguity before final response generation. The framework operates between the patient and an otherwise unchanged downstream language model. It interprets the initial query, uses a task-specific knowledge graph to construct a set of plausible hypotheses, identifies the missing patient-context variables needed to distinguish among them, and asks targeted follow-up questions. The original query and the acquired context are then combined into a clarified prompt for the downstream model. We evaluated the framework across five language models using two controlled ambiguity-mitigation benchmarks: diagnosis retrieval from 1,034 symptom queries with clinically relevant evidence systematically masked, and dietary-safety classification from 487 queries with decisive health context omitted. The framework was compared with direct answering of the underspecified query and with rephrasing the same query without acquiring new patient information. In diagnosis retrieval, it increased overall exact Top-1 accuracy by at least 57.1 percentage points and selective exact Recall@5 by at least 77.7 percentage points across the five evaluated models compared with direct prompting. In dietary-safety classification, it improved accuracy across all five models and achieved the highest Matthews correlation coefficient for four…
[NLP-25] PolicyGuide: From Guarding One Action to Guiding the Whole Workflow for Policy-Compliant LLM Agents
【速读】： 该论文旨在解决大语言模型（Large Language Model, LLM）在客户服务场景中因违反组织政策而导致的合规性问题，具体表现为两类风险：一是执行被禁止的操作（如为不符合条件的用户办理变更），二是遗漏必要的流程要求（如身份验证或确认步骤）。现有方法存在局限性：运行时防护机制虽可干预高风险行为，但仅基于局部动作判断，无法指导多步骤流程中的行为；而工作流遵循系统虽能保障流程完整性，却侧重于流程完成而非对代理行为进行持续合规性约束。为此，论文提出PolicyGuide框架，其核心创新在于将各领域政策转化为显式的流程图（workflow graph），并在用户交互轮次边界处调用主动验证器（proactive verifier）。该验证器基于持久化的图状态，动态协调未完成请求，并提供针对具体步骤的合规修复建议，引导代理沿符合政策的路径推进。实验结果表明，在包含航空、零售和电信三个领域的τ²-bench基准测试中，采用GPT-5.4模型的PolicyGuide将平均通过率（mean Pass⁴）从0.42提升至0.62，其中电信领域提升最为显著（由0.19增至0.61），因其流程结构最复杂。此外，所构建的工作流具备跨模型泛化能力，可成功迁移至Claude Sonnet 4.6与Gemini 2.5 Pro。补充评估显示，该方法在对抗性用户攻击下表现出最低的成功率，且在人工设计的工作流级合规性验证中展现出最强的程序遵循能力。
链接: https://arxiv.org/abs/2608.19861
作者: Seongjae Kang,Taehyung Yu,Sung Ju Hwang
机构: KAIST(韩国科学技术院); DeepAuto.ai
类目: Artificial Intelligence (cs.AI); Computation and Language (cs.CL); Machine Learning (cs.LG)
备注: 26 pages, 15 figures, including appendices
Abstract:Customer-service LLM agents must follow organizational policy when acting on a user’s behalf. Compliance failures arise from either forbidden actions, such as granting an ineligible change, or omitted procedural requirements, such as identification or confirmation. Runtime safeguards can intervene on risky actions, but action-local checks do not guide an agent through a multi-step procedure. Workflow-following systems support prescribed process execution, but primarily target workflow completion rather than safeguarding agent behavior. PolicyGuide instead compiles each domain policy into a workflow graph and invokes a proactive verifier at user-turn boundaries. From persisted graph state, the verifier reconciles open requests and returns step-specific remediation along a policy-compliant path. Across the \tau^2 -bench airline, retail, and telecom domains with a GPT-5.4 agent and verifier, PolicyGuide raises mean \mathrmPass^4 from 0.42 to 0.62 , with the largest gain on telecom ( 0.19 to 0.61 ), the most workflow-structured domain. The same workflows transfer to Claude Sonnet 4.6 and Gemini 2.5 Pro agents. Complementary evaluations find the lowest observed attack-success rate under adversarial users and the strongest procedural compliance in an author-designed workflow-level validation.
[NLP-26] owards Clinically Faithful Medical Image Captioning via Enhanced Vision-Language Alignment
【速读】： 该论文旨在解决医学图像描述生成中临床可靠性不足的问题，尤其针对灰度成像模式、细微解剖线索、专业医学术语表达以及数据质量差异等挑战。其核心解决方案在于区分并强化训练时与推理时的临床对齐机制：在训练阶段引入基于临床相关奖励的MedPAIR-SCST方法，通过强化学习优化生成分布以提升临床一致性；在推理阶段采用单嵌入重排序策略，从候选描述中筛选出最符合临床概念空间的最优结果。该框架融合BioMedCLIP与SigLIP2双视觉编码器、Q-Former及LLaMA解码器，并引入UMLS概念/类型预测作为辅助任务，从而增强模型对临床语义信息的捕捉能力。实验表明，多编码器设计与概念级辅助学习有助于保留关键临床信息，而推理阶段的重排序与训练阶段的强化学习协同作用，可在数据受限条件下显著提升生成描述的语义与临床对齐性，推动更可信的医学图像描述生成。
链接: https://arxiv.org/abs/2608.19825
作者: Yunseo Lee,Hyun Jun Kim,Heeseung Shin,Changwon Lim
机构: Chung-Ang University (中央大学)
类目: Computer Vision and Pattern Recognition (cs.CV); Computation and Language (cs.CL)
备注: 10 pages, 2 figures, 7 tables. Preprint submitted to IEEE for possible publication
Abstract:Medical image captioning is a technique that accelerates early-stage diagnostic workflows and enhances the interpretability of medical diagnostic AI systems. However, unlike general image captioning, clinically reliable captioning remains challenging due to grayscale-based modalities, subtle anatomical cues, specialized medical phrasing, and variations in data quality. Despite recent advances in large vision-language models, fluent outputs do not necessarily guarantee sufficient alignment with clinical concept spaces or evaluation criteria. To address this issue, we propose a framework that strengthens clinical alignment by separating and enhancing training-time alignment and inference-time alignment. We build a medical image captioning pipeline that integrates single/dual vision encoders based on BioMedCLIP and SigLIP2, a Q-Former, and a LLaMA-based decoder, and examine the contribution of auxiliary learning for UMLS concept/type prediction. At inference, we apply single-embedding-based reranking to select the best caption among candidates, while at training we introduce MedPAIR-SCST, which combines clinically relevant rewards to shift the generative distribution toward improved clinical alignment. Our experiments show that complementary visual representations with a multi-encoder design and concept-level auxiliary learning help preserve clinically meaningful information. Furthermore, inference-time reranking provides a practical way to improve semantic and clinical alignment without additional training, whereas MedPAIR-SCST goes beyond selection by directly improving the model’s distribution to generate more consistent and clinically grounded captions. These findings suggest that jointly leveraging selection-based alignment and reinforcement-learning-based alignment can promote more trustworthy medical image captioning even in data-constrained settings.
[NLP-27] MileGPO: Milestone Inference with Local Evidence for Graph-Based Policy Optimization of Long-Horizon LLM Agents
【速读】： 该论文旨在解决长时域智能体强化学习中因仅依赖最终奖励而导致的信用分配（credit assignment）难题，尤其针对现有方法在轨迹层面信号分解为步骤级信用时易忽略有意义中间里程碑的问题。其核心解决方案是提出一种名为MileGPO（基于图的策略优化中的里程碑发现与局部证据）的方法，通过三个关键设计实现过程级信用的精准推断：首先，通过成功轨迹上的候选里程碑发现与失败轨迹上的重复陷阱识别来定位潜在里程碑；其次，引入可靠性校准的奖励塑造（Reliability-Calibrated Shaping, RCS），基于结果置信度对候选里程碑进行加权，强化高可靠性里程碑与陷阱，抑制不确定性高的信号；最后，采用进展对比校准（Progress-Contrastive Calibration, PCC），检验候选里程碑是否反映局部进展，并通过同状态分支间的替代路径比较验证其有效性。该方法无需额外辅助模型或环境交互，实验在ALFWorld和WebShop上均取得当前最优性能，且在ALFWorld上表现出较小的分布内到分布外泛化差距。消融研究与信用诊断表明，可靠性加权、局部进展判别及同状态分支证据三者协同补充了里程碑发现机制，有效缓解了中间信用判断的模糊性问题。
链接: https://arxiv.org/abs/2608.19803
作者: Bo Qian,Yuting Wu,Shuang Zeng,Huaiyu Wan,Dalin Zhang,Jiqiang Liu
机构: Beijing Jiaotong University (北京交通大学)
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Computation and Language (cs.CL)
备注:
Abstract:Credit assignment is challenging in long-horizon agentic reinforcement learning, where supervision often comes only from final rewards. Existing methods refine trajectory-level signals into step-level credits through step grouping or graph-based advantage estimation, but can overlook meaningful intermediate milestones. We propose MileGPO (Milestone Inference with Local Evidence for Graph-Based Policy Optimization), which derives process-level credit from grouped on-policy rollouts through three designs. Milestone Discovery identifies candidate milestones on successful rollouts and recurring traps on failed ones. Reliability-Calibrated Shaping (RCS) weights these candidates by outcome-based confidence, strengthening reliable milestones and traps while down-weighting uncertain ones. Progress-Contrastive Calibration (PCC) further tests whether a candidate reflects local progress and whether its incoming ansition outperforms observed alternatives from the same this http URL requires neither auxiliary models nor additional environment interaction. Experiments on ALFWorld and WebShop show state-of-the-art performance and a small in-distribution to out-of-distribution gap on ALFWorld. Ablations and credit diagnostics indicate that reliability weighting, local progress, and same-state branch evidence complement milestone discovery and resolve ambiguous intermediate credit.
[NLP-28] Stopping and Routing LLM Judge Panels
【速读】： 该论文旨在解决大语言模型（Large Language Model, LLM）评估流程中多候选评判者（judge）部署的复杂决策问题，即如何在不同样本上选择合适的评判者、何时停止构建评判小组，以及如何权衡成本与评估质量。其核心挑战在于传统方法仅关注“哪个评判者最优”，而忽视了评判者组合策略的动态优化。解决方案的关键在于将评判小组设计建模为一种角色条件分配问题（role-conditioned allocation problem），基于少量标注审计数据、声明的样本子集（declared slices）及评判者成本，推断出各类评判者的相对角色：冗余副本（copies）不提供条件信息，互补者（complements）可提升整体面板性能，专业评判者（specialists）仅在特定子集上有价值。由此导出的策略为：剔除冗余副本，全局引入互补者，按条件路由专业评判者，并在验证收益低于阈值时终止。实验覆盖推理、代码、安全、偏好、奖励模型、摘要和数学等多个评估场景，结果生成了一套可复用、可审计的评判调用方案，形成“评判调用制度图谱”——包括对可部署子集路由专业评判者、在饱和验证环境下及时停止、保留高风险-收益比的广泛集成组，以及忽略条件性副本。该方法显著优于单个评判者、静态小组、多样性启发式、全量堆叠、可靠性裁判团及节俭级联等基准方案。
链接: https://arxiv.org/abs/2608.19802
作者: Bin Zhu,Yi Xie,Yanghui Rao
机构: Sun Yat-sen University (中山大学)
类目: Computation and Language (cs.CL)
备注: 21 pages, 2 figures, 20 tables. Accepted at WISE 2026
Abstract:LLM evaluation pipelines often have many candidate judges: general LLM-as-a-judge prompts, reward models, safety classifiers, confidence variants, and task-specific verifiers. The deployment question is not only which judge is best, but which judges should be called, on which examples, and when panel construction should stop. We formulate judge-panel design as a role-conditioned allocation problem. From a small labeled audit set, declared slices, and judge costs, the method estimates target-relative roles: copies add no conditional information, complements improve the global panel, and specialists help only on slices. These roles induce a policy: drop copies, add complements globally, route specialists conditionally, and stop when validation gain falls below a threshold. Across reasoning, code, safety, preference, reward-model, summarization, and math audits, the method is compared with single judges, flat panels, matched diversity heuristics, full-call stacking, reliability juries, and frugal cascades. The result is a regime map for judge calls: route specialists on deployable slices, stop in saturated verifier regimes, keep broad ensembles when their risk benefit is worth the cost, and ignore conditional copies. The output is a reusable, auditable call plan for the next evaluation batch.
[NLP-29] LoRA-GA2: Low Rank Adaptation with Multi-step Gradient Adaptive Alignment
【速读】： 该论文旨在解决低秩适配（LoRA）与全量微调之间持续存在的性能差距问题。尽管LoRA通过降低内存开销实现了高效的模型微调，但其在拟合全量微调更新方向上的能力受限，主要源于对梯度动态信息的不完整捕捉。现有方法虽尝试通过单步梯度近似来对齐主方向或内在维度，但仍无法充分反映梯度演化的复杂性。为此，本文提出LoRA-GA²，一种充分利用多步梯度信息的高效微调算法。其关键创新在于引入一个轻量级探针（lightweight probe），用于捕获预训练权重的多步梯度信息，该设计不增加额外GPU显存占用，仅带来可忽略的时间开销；同时，基于多步梯度构建谱感知（spectrum-aware）、重要性驱动的秩分配策略及最优初始化方案，显著提升了参数更新的有效性。实验结果表明，LoRA-GA²在保持原始LoRA高效率的同时，在GLUE、GSM8K和HumanEval等多个基准上均显著超越现有LoRA变体，平均性能提升分别达0.66、1.03和0.87分，充分验证了其在捕捉梯度动态与优化更新路径方面的有效性。
链接: https://arxiv.org/abs/2608.19800
作者: Haonan He,Xinyue Fan
机构: 未知
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
备注:
Abstract:Low-Rank Adaptation (LoRA) is a prominent fine-tuning method for large models, achieving competitive performance with reduced memory overhead. However, a persistent performance gap remains between LoRA and full fine-tuning. Recent studies have sought to narrow this gap by employing one-step gradient approximations of pretrained weights to align LoRA updates with the principal directions or intrinsic dimensionalities of full fine-tuning updates. Nevertheless, these approaches fail to capture the full dynamics of the gradients. In this paper, we propose LoRA-GA ^2 , an effective fine-tuning algorithm that fully leverages multi-step gradient information. Specifically, we introduce a lightweight probe for multi-step gradients of pretrained weights that incurs no additional GPU memory cost and only marginal time overhead. We further employ a spectrum-aware, importance-based rank allocation and optimal initialization derived from multi-step gradients. Extensive experimental results demonstrate that LoRA-GA ^2 consistently outperforms existing LoRA variants while preserving the efficiency advantages of vanilla LoRA. For instance, LoRA-GA ^2 surpasses the leading baseline by an average of 0.66 points on the GLUE benchmark, and outperforms the strongest baseline by 1.03 points on GSM8K and 0.87 points on HumanEval, respectively.
[NLP-30] SWE-bench Science: Can Coding Agents Resolve Engineering Tasks in Science?
【速读】： 该论文旨在解决科学软件工程中编码代理（coding agents）在修复科学代码时失败原因不明确的问题，尤其关注其在真实科研项目场景下的表现与局限性。现有评估多聚焦于任务整体成功率，缺乏对失败机制的深入分析。为此，研究提出SWE-bench Science——一个基于仓库级别的基准测试集，涵盖来自98个GitHub仓库、横跨20个科学领域的119项任务，按问题驱动（Issue-driven）、专家探索（Expert-exploratory）和工程集成（Engineering-integration）三类范式组织。实验表明，即使表现最优的Claude Code with Opus-5（max）在pass@1指标上也低于50%，揭示了科学软件工程任务的高度复杂性。研究识别出四类核心失败机制：科学知识或抽象能力不足、错误探索或浅层修复、修复覆盖不全或系统集成缺失、以及无法将科学知识泛化至未见案例。进一步的配对消融实验显示，显式的科学指导并非始终有益：合理且准确的知识可约束修复路径、提升平均性能与令牌效率，而偏差较大的引导则可能引发锚定效应，反而降低精确修复成功率。因此，该研究的关键贡献在于构建了一个全面反映科学软件工程挑战的基准，并揭示了科学知识在代码修复中的双重作用——既可赋能又可误导，为未来编码代理的设计提供了重要洞见。
链接: https://arxiv.org/abs/2608.19799
作者: Zhipeng Xu,Jiahao Lu,Yining Zheng,Yuxin Wang,Xipeng Qiu
机构: 未知
类目: Computation and Language (cs.CL); Software Engineering (cs.SE)
备注: 26 pages, 7 figures
Abstract:Software increasingly functions as part of the scientific instrument itself, making failures in scientific code capable of compromising not only program behavior but also the evidence underlying scientific conclusions. Yet existing evaluations of coding agents largely emphasize aggregate task success, providing limited insight into why agents fail when repairing scientific software. We introduce \textbfSWE-bench Science, a repository-level benchmark for scientific software engineering comprising 119 tasks from 98 GitHub repositories across 20 scientific domains. Each task is organized into one of three paradigms: Issue-driven, Expert-exploratory, and Engineering-integration. Even the best-performing agent, \textbfClaude Code with Opus-5 (max), achieves a pass@1 below 50%, highlighting the substantial challenges posed by scientific software engineering. We identify four recurring failure mechanisms: deficits in scientific knowledge or abstraction, misguided exploration or surface-level repair, incomplete repair coverage or system integration, and failures to generalize scientific knowledge beyond observed cases in our analysis. We further conduct a paired ablation that removes explicit scientific guidance while preserving the repository and executable engineering context. The results show that scientific knowledge is not uniformly beneficial: well-grounded information can constrain repair and improve average performance and token efficiency, whereas poorly aligned guidance can induce anchoring and does not necessarily improve exact repair success. Together, SWE-bench Science provides a broad testbed for studying both the capabilities and failure mechanisms of coding agents in scientific software engineering.
[NLP-31] Credit Without Ground Truth: Auditing Step-Level Credit Assignment in LLM Agents Against Executed Replay
【速读】： 该论文旨在解决当前大语言模型（LLM）智能体在单智能体工具环境（ALFWorld）中训练时，用于分配步骤级信用信号（step-level credit signals）的各类方法——包括基于大模型判断的评分（LLM-judge scores）、基于结果条件化的对数概率比（outcome-conditioned logprob ratios），以及策略自身置信度——均无法有效识别出真正对最终结果具有因果影响的决策步骤这一关键问题。其核心挑战在于，现有评估体系依赖人工标注的“步骤正确性”作为基准，而本文提出应以“步骤贡献度”（step contribution）为因果真值进行审计，即通过重采样策略在每个决策点的替代动作并向前推进模拟，来衡量实际对结果的影响。研究发现，因果贡献具有稀疏性（仅30.5%的决策点存在可测量效应），且可测性高度依赖于具体模型表现，不同规模相近的策略间无支持反事实的决策点比例差异达两倍（13.1% vs. 26.8%）。进一步分析揭示，现有信用信号的失败根源在于：其隐含的信用传递机制主要反映的是策略的语言流畅性（中位秩相关性+0.75），而非因果有效性；而基于结果的条件化并未引入额外的因果信息（偏相关性-0.004，Qwen模型）。尽管仅依赖置信度的路由机制无法超越随机水平识别关键步骤，但能将评判成本降低13.1%（每回合）或14.0%（每轨迹）。在一项七臂预注册训练实验中，所有干预组均未显著优于未经训练的基线策略，且所谓“信用规则”的表现差异可完全由训练剂量解释——更稀疏的信用信号因保留样本量更少，导致优化器步数存在数量级差异，而非信用内容本身的有效性。因此，该研究强调，比较不同信用规则必须匹配有效样本量，否则所衡量的仅为训练剂量效应，而非信用机制的真实性能。
链接: https://arxiv.org/abs/2608.19760
作者: Haiyue Zhang
机构: University of Southern California(南加州大学); Los Angeles, CA, USA
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Computation and Language (cs.CL)
备注: 49 pages, 7 figures. Pre-registered; frozen analysis plans and prompts included in the appendices. Under review
Abstract:Audited against causal ground truth from executed replay in a single-agent tool environment (ALFWorld), none of the step-level credit signals used to train LLM agents – LLM-judge scores, outcome-conditioned logprob ratios, or the policy’s own confidence – identifies which steps causally matter better than chance. Existing evaluations grade these signals against annotated step correctness; we audit them against step contribution – what re-sampling the policy’s own alternatives at each decision point and rolling forward actually changes about the outcome – and the two come apart. The ground truth itself is structured: causal contribution is sparse (30.5% of decision points where ground truth is defined carry measurable effect), and measurability is model-dependent – the fraction of points with no policy-supported counterfactual differs by a factor of two (13.1% vs. 26.8%) between two similar-scale policies. The failure mode is identifiable: implicit credit echoes the policy’s fluency (median rank correlation +0.75, replicating at +0.70 in a second family under a corrected instrument), while conditioning on the outcome adds no causal information (partial correlation -0.004, Qwen). A confidence-only router recovers pivotal steps at chance level, but cuts judge cost by 13.1% per turn (14.0% per trajectory). In a seven-arm pre-registered training experiment, no arm reliably outperforms the untrained policy, and the checkpoints’ apparent instrument signature is fully explained by training dose – sparser credit retains fewer examples, an order-of-magnitude spread in optimizer steps – not credit content. Comparisons of credit rules must therefore match effective sample size, or they measure dose, not credit.
[NLP-32] FlashPrefill V2: Block-Sparse Prefill Attention for Long-Context LLM Serving
【速读】： 该论文旨在解决大语言模型在长上下文建模中因注意力机制的二次复杂度带来的计算瓶颈，尤其是在资源密集型预填充（prefilling）阶段。其核心挑战在于如何在保持高精度的同时实现高效的稀疏化注意力计算，以支持实际生产环境中的长序列推理。解决方案的关键在于提出FlashPrefill V2，通过三个关键改进实现从原型到实用部署的跃迁：首先引入均值修正项，有效抑制近似误差，在极端稀疏条件下仍能控制性能下降；其次重构稀疏注意力算子，集成PackGQA内存访问、线程束（warp）特化及乒乓流水线（pingpong pipelining），与最新的FlashAttention-3/4实现对齐，并支持FP8量化以满足工业级推理需求；最后原生支持分页键值缓存（paged KV cache）和连续批处理（continuous batching），可无缝集成至SGLang等现代推理框架。在NVIDIA H20 GPU上的大量实验表明，FlashPrefill V2在128K上下文长度下，相比FlashAttention-2分别实现了高达47.26倍（FP8精度）和27.19倍（BF16精度）的加速，且在FP8下仍较对齐FlashAttention-3/4的稠密基线提升30.49倍，显著推动了长上下文生成式AI (Generative AI) 的高效服务落地。
链接: https://arxiv.org/abs/2608.19758
作者: Qihang Fan,Huaibo Huang,Zhiying Wu,Bingning Wang,Ran He
机构: CASIA(中国科学院自动化研究所); UCAS(中国科学院大学); WeChat(微信)
类目: Computation and Language (cs.CL)
备注: FlashPrefill V2
Abstract:Long-context modeling is a pivotal capability for Large Language Models, yet the quadratic complexity of attention remains a critical bottleneck, particularly during the compute-intensive prefilling phase. Our previous work, FlashPrefill, mitigates this cost through instantaneous pattern discovery and max-based dynamic thresholding; however, it remains an algorithmic prototype that is still distant from production deployment. In this paper, we present FlashPrefill V2, which evolves FlashPrefill from a prototype toward practical long-context serving along three dimensions. First, we introduce a mean correction term that effectively suppresses the approximation error, keeping performance degradation manageable even at extreme sparsity levels. Second, we redesign the sparse attention operator with PackGQA memory access, warp specialization, and pingpong pipelining, fully aligning with the latest FlashAttention-3/4 implementations and supporting FP8 inference to meet practical quantization requirements. Third, FlashPrefill V2 natively supports paged KV cache and continuous batching, allowing integration as an attention backend in modern inference frameworks such as SGLang. Extensive evaluations on NVIDIA H20 GPUs—among the most widely deployed inference accelerators—demonstrate that FlashPrefill V2 delivers up to 47.26x and 27.19x speedups over FlashAttention-2 at 128K context length under FP8 and BF16 precision, respectively, and, in FP8, still achieves a 30.49x speedup against an FA3/4-aligned dense baseline.
[NLP-33] runcate Bad Upweight Good: BoN-Style Distillation via Rank-Based Classification
【速读】： 该论文旨在解决生成式AI（Generative AI）中基于奖励模型的推理时选择方法（如Best-of-N）在实际应用中效率低下的问题。现有方法通过蒸馏将多候选采样与选择过程转化为单一策略，但其依赖的基于排名的重加权策略通常采用平滑的全支持重加权，导致低排名生成结果虽被弱化但仍保留在分布支持中，无法有效剔除劣质输出。为此，本文提出TUP（Truncate-bad, Upweight-good Policy）：一种先移除低排名生成结果（下尾截断），仅对保留的上尾部分进行可调锐度重加权的策略。其关键创新在于引入下尾截断机制，从而避免对劣质样本的冗余建模，并实现无需提示依赖的闭式归一化，支持完全离线训练，利用偏移截断胜率作为软标签、蒸馏后的参考对数似然比作为逻辑值，通过二元交叉熵优化。理论上，在特定假设下，证明了对于任意未知的最优奖励函数，最优单调排名重加权可由下尾截断规则实现，为剔除低质量生成结果提供了形式化依据；实证结果表明，TUP在性能上可媲美先进的离线对齐基线方法。
链接: https://arxiv.org/abs/2608.19748
作者: Yarin Bar,Yaniv Romano
机构: Technion–Israel Institute of Technology(以色列理工学院)
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Computation and Language (cs.CL)
备注:
Abstract:Inference-time selection methods, such as Best-of-N, improve generation by sampling a pool of candidates and selecting the top-ranked completion according to a reward model. Distillation seeks to amortize this procedure into a single policy by replacing raw rewards with in-pool ranks and learning a policy that upweights higher-ranked completions. However, existing rank-based policies typically use smooth full-support reweighting, so low-ranked completions receive less mass but remain in the target support. Although a sharper reweighting reduces lower-tail mass, it also increases reliance on brittle ranking at the top made by a single reward model. We propose TUP: a Truncate-bad, Upweight-good Policy that removes low-ranked completions from the support and reweights only the retained upper tail with a tunable sharpness. TUP admits a closed-form, prompt-independent normalization and can be trained fully offline via binary cross-entropy, using shifted-truncated win-rates as soft labels and distilled-to-reference log-likelihood ratios as logits. Theoretically, under certain assumptions, we show that for any unknown oracle reward, the best monotone rank-reweighting can be matched by a lower-tail truncation rule, providing formal support for removing the lower tail rather than merely downweighting it. Empirically, we show that TUP is competitive with strong offline alignment baselines.
[NLP-34] PersonalBench: Measuring the Authorship Gap in LLM Personalization
【速读】： 该论文旨在解决当前个性化文本生成评估中缺乏对生成内容是否真实贴近目标作者写作风格的量化衡量问题。现有基准多关注任务准确性或偏好对齐，而忽视了对作者风格相似性的实质检验。为此，作者提出PersonalBench这一基准，通过三个独立评估维度——基于训练的作者身份验证模型（LUAR）、大语言模型作为裁判（LLM-as-judge）以及自动化文体学分析——系统评估推理阶段个性化方法的效果。其核心发现在于：尽管个性化方法能生成在风格上可区分的目标作者文本（LUAR AUC=0.918），但这些生成文本与真实人类作者之间的风格相似性始终低于人类间自然差异的下限（实际相似度范围为0.484–0.508，而跨作者人类间相似性最低为0.626），表明当前方法无法跨越人机风格鸿沟。更关键的是，所有方法在LUAR指标上的表现几乎无统计差异（方差仅0.024），尽管在LLM裁判视角下看似存在差异，这种表观差异源于特征提取与风格画像构建之间的循环依赖。研究进一步验证了LUAR在本语料库中的可靠性（单篇AUC=0.76，多篇AUC=0.96）。最终，论文将PersonalBench公开为一个校准化的评估工具，揭示当前推理阶段个性化技术虽可调节大模型风格，但尚未能真正实现向人类作者风格的逼近。
链接: https://arxiv.org/abs/2608.19746
作者: Yash Ganpat Sawant
机构: 独立人工智能研究员(Independent AI Researcher)
类目: Computation and Language (cs.CL)
备注: 17 pages. Extended version
Abstract:Personalized text generation aims to make LLMs write in a specific individual’s style, yet existing benchmarks measure task accuracy or preference alignment rather than whether the model’s output actually resembles the target author’s writing. We introduce PersonalBench, a benchmark that evaluates inference-time personalization methods through three independent lenses: LUAR (a trained authorship verification model), an LLM-as-judge, and automated stylometrics. Across 50 authors, 1,000 generations, and two model families (Qwen 3, GLM-4), we find that personalization methods do produce author-differentiated output (LUAR discriminates target authors within generated text at AUC=0.918) but this differentiation never crosses the human-LLM boundary. All methods achieve LUAR similarity to real authors in the range 0.484-0.508, below the cross-author human floor of 0.626 (ceiling 0.756). The LLM’s own authorship fingerprint dominates: generated text is more distant from any human author than random humans are from each other. Methods are statistically indistinguishable from each other on LUAR (spread 0.024) despite appearing differentiated on the LLM judge, a discrepancy we trace to circularity between trait extraction and profile extraction. We validate that LUAR reliably measures authorship in our corpus (AUC=0.76 single-post, 0.96 multi-post). We release PersonalBench as a calibrated measuring stick: inference-time personalization modulates the LLM’s style but does not bridge the gap to human authorship.
[NLP-35] One Success Isnt Reliability: Thinkingbox a Sandbox and Benchmark for Agents in Stateful Business Workflows
【速读】： 该论文旨在解决当前生成式智能体（Agent）评估体系中缺乏对复杂、状态依赖型任务的精准衡量问题，尤其在涉及多轮交互、工具协调、领域政策遵循及持久状态正确转换的实际业务场景中，现有基准难以有效区分智能体是否真正完成任务。其核心挑战在于：仅通过生成合理响应或合法工具调用无法反映端到端任务的成功与否，而真实业务流程往往要求智能体在多轮交互中持续获取缺失信息、遵守规则、协同多个工具，并确保状态转移的准确性与无副作用。为此，论文提出Thinkingbox——一个支持工具-智能体-用户交互的沙箱环境，具备MCP兼容的独立工具会话、完整的执行轨迹记录以及基于终端后端状态的结果评估能力。在此基础上构建的Thinkingbox-bench包含507个受策略约束的工作流，覆盖零售、酒店、汽车保险、数字银行内部IT及咨询类IT/HR支持等真实业务场景。评估采用任务特异性的可执行检查机制，严格验证轨迹有效性并排除错误、遗漏或冗余操作；部分任务还额外校验最终响应的属性。实验表明，即使在最强模型下，pass@1仅为65.36%，而pass^20仅达25.25%，且大量失败案例表现出干净终止和有效的状态变更动作，揭示了传统以响应或工具调用层面信号作为代理指标的局限性。结果表明，当前智能体在偶然找到成功路径方面表现尚可，但在可靠完成具有状态依赖性的实际业务任务方面仍存在显著差距。研究团队已开源Thinkingbox与Thinkingbox-bench。
链接: https://arxiv.org/abs/2608.19741
作者: Zhuochun Li,Youngmin Ko,Ali Keramati,Nicola Ferri,Susana Palmaz Lopez Pelaez,Liang-Chun Tsai,Calvin Wang,Mirco Milletari,Tuhin Kundu,Vadim Smolyakov,Kjartan Olafsson,Tommy Guy
机构: University of Pittsburgh (匹兹堡大学); Northwestern University (西北大学); University of California, Irvine (加州大学欧文分校); Microsoft (微软)
类目: Computation and Language (cs.CL); Databases (cs.DB)
备注:
Abstract:Recent agent benchmarks increasingly ground evaluation in executable environments, from code repair to web navigation, app APIs, and function calling. Yet completing consequential work beyond code requires more than producing a plausible response or valid tool call: agents must gather missing information over multiple turns, follow domain policies, coordinate dependent tools, and realize the correct persistent state transition without collateral effects. In this paper, we introduce Thinkingbox, a sandbox for tool-agent-user interaction that provides isolated MCP-compatible tool sessions, complete execution traces, and outcome evaluation over terminal backend state. Built on this sandbox, Thinkingbox-bench contains 507 policy-conditioned workflows across numerous scenarios, including retail, hospitality, auto insurance, neobank internal IT, and consulting IT/HR support. Each attempt is evaluated by task-specific executable checks that accept valid trajectories while rejecting wrong, missing, or extra effects; designated tasks additionally check required properties of the final response. Across proprietary and open-weight models, the strongest achieves 65.36% pass@1, but only 25.25% pass^20. Moreover, many failed trials show clean termination and valid state-changing actions, showing that response or tool-call-level signals are not clear proxies for end-to-end task completion. Thinkingbox-bench reveals a large gap between occasionally finding a successful trajectory and reliably completing stateful business tasks. We release both Thinkingbox and Thinkingbox-Bench: this https URL
[NLP-36] mpJail: Temporal Jailbreak Attack against Large Vision-Language Models via Subtitle Scheduling
【速读】： 该论文旨在解决当前大型视觉语言模型（Large Vision-Language Models, LVLMs）在视频理解与推理任务中面临的视频越狱攻击（video jailbreak）问题，尤其是针对现有方法仅关注视频中嵌入文本内容的语义而忽视其时序组织方式的局限性。研究表明，越狱攻击的有效性不仅取决于文本内容的语义，更受其时间呈现特性的影响，包括持续时间与时间槽分配等关键因素。为此，论文提出一种名为TempJail的黑盒视频越狱框架，其核心在于利用真实视频中普遍存在的字幕作为攻击媒介——字幕兼具语义表达能力与精确的时间控制优势，且不会造成视觉上的显著干扰。TempJail通过构建与目标查询对齐的对话式字幕序列，并优化其时间调度策略，以系统性地挖掘LVLM在时序处理方面的潜在漏洞，从而诱导模型生成满足恶意意图的响应。实验结果表明，在四个代表性LVLM及两个数据集上的评估中，TempJail在所有测试场景下均达到最高攻击成功率，相较于最强基线，在GPT-5和Gemini 3.5-Flash上分别提升了53和18个百分点的平均攻击成功率（ASR），验证了时序控制在视频越狱攻击中的决定性作用。
链接: https://arxiv.org/abs/2608.19737
作者: Ling Zhou,Yihao Huang,Jingling Sun,Zhiwen Tian,Yi Zeng,Qihe Liu,Shijie Zhou
机构: University of Electronic Science and Technology of China (电子科技大学); East China Normal University (华东师范大学)
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI); Computation and Language (cs.CL)
备注: 8 pages,4 figures
Abstract:Large vision-language models (LVLMs) have achieved remarkable progress in video understanding and reasoning. Despite extensive studies on text- and image-based jailbreaks, video jailbreaks against LVLMs remain largely unexplored. Existing video jailbreak methods mainly manipulate textual content embedded in videos, while overlooking how such information is organized over time. Our analysis reveals that jailbreak effectiveness depends not only on the semantics of textual information but also on its temporal presentation, including duration and timing-slot allocation. Motivated by this finding, we use subtitles, which are common in real-world videos and allow semantic content to be presented under precise temporal control without appearing visually intrusive, as a natural attack medium. Based on this insight, we propose TempJail, a black-box video-based jailbreak framework that constructs query-aligned dialogue-style subtitle sequences and optimizes their temporal scheduling to exploit temporal vulnerabilities in LVLMs and elicit responses that satisfy the harmful intent of the source query. Extensive experiments on four representative LVLMs and two datasets demonstrate that TempJail achieves the highest attack success rate across all evaluated model–dataset settings, outperforming the strongest baseline by 53 and 18 percentage points in dataset-averaged ASR on GPT-5 and Gemini 3.5-Flash, respectively.
[NLP-37] Projector Is All You Train
【速读】： 该论文旨在解决多模态大语言模型（Multimodal Large Language Model, MLLM）在适配新模态时是否必须对语言模型主干（backbone）进行微调这一关键问题。传统训练流程通常同时微调语言模型主干与连接主干和特定模态编码器之间的投影模块（projector），但本文通过在3D MLLM上的实验发现，仅微调投影模块即可实现与联合训练模型相当甚至更优的多模态性能，且显著避免了联合训练导致的语言模型原有能力退化（即“灾难性遗忘”或能力漂移）。其解决方案的关键在于：仅训练投影模块即可有效实现跨模态对齐，而无需更新语言模型主干参数。该方法不仅提升了训练效率（样本吞吐量约为联合训练的两倍），还在多种基准测试（包括3D分类、图像描述生成以及评估语言、视觉和空间推理能力的标准任务）中验证了其有效性与泛化性。
链接: https://arxiv.org/abs/2608.19726
作者: Nyx Iskandar,Saathvik Selvan,Slater Victoroff
机构: Ramen VR; University of California, Berkeley (加州大学伯克利分校); iph.so
类目: Computation and Language (cs.CL); Computer Vision and Pattern Recognition (cs.CV); Machine Learning (cs.LG)
备注:
Abstract:The typical training process of a multimodal large language model (MLLM) involves adapting both the language model backbone and the projector between the backbone and a modality-specific encoder. We ask whether fine-tuning the backbone of an MLLM is necessary to adapt it to a new modality. Through experiments on 3D MLLMs, we find that training only the projector is sufficient to achieve strong multimodal performance relative to existing baseline models and our jointly trained MLLMs with the same encoder and backbone. We also show that joint training leads to undesirable drift in existing capabilities of the language model, which projector-only training avoids by definition. Furthermore, projector-only training has approximately twice the training sample throughput of joint training. We validate our findings across different language model backbones via 3D classification and captioning benchmarks as well as standard benchmarks evaluating language, vision, and spatial reasoning capabilities.
[NLP-38] StreamSoccer: Event-Driven Memory for Streaming Soccer Commentary
【速读】： 该论文旨在解决流式视频理解中如何在有限计算与内存约束下，实现对持续输入视频流的因果状态更新，并将不断增长的历史信息组织为可演化、可持久化且可召回的语义单元这一核心挑战。尤其针对实时足球解说场景，系统需在每一时刻仅依赖此前可用信息，完成对已完成事件的描述、近期比赛的总结、早期事件的回忆或保持沉默。其解决方案的关键在于提出一种基于事件记忆（event memory）的流式架构——StreamSoccer，通过固定预算的主动记忆机制整合视频流，将已完成事件的状态局部保留并结构化为可检索的历史记录；同时采用统一生成器结合当前、近期与历史上下文，支持多模式解说生成，并由规则辅助调度器动态选择输出模式或静默。与传统以帧、视觉标记或缓存为中心的流式视频-语言模型，以及依赖预定义片段或输出时间戳的足球解说方法不同，StreamSoccer显式建模了事件生命周期，实现了跨时间尺度的语义组织。实验构建了三轨流式足球解说数据集与分层评估协议，在多个参考锚点上，其在当前事件、近期窗口和历史记忆任务上的CIDEr得分分别达到38.62、23.96和17.39，分别排名第一和第二，且消融实验表明局部事件状态显著提升各任务性能，完整系统表现最优。在58场比赛共174次原始视频运行中，每分钟推理延迟的第95百分位（RTF p95）维持在0.10至0.22之间，未随比赛进程呈现持续增长，验证了事件记忆机制在控制长时历史计算开销的同时，有效支撑跨时间尺度的流式解说能力。
链接: https://arxiv.org/abs/2608.19723
作者: Chenxi Shao,Bozhong Wang,Jiaxin Huang,Zhao Liu,Sunwei Zhu,Tianxin Hang,Gaoqi He,Yang Li,Changbo Wang
机构: Migu Video Technology Co., Ltd., Shanghai, China; South China University of Technology, Guangzhou, Guangdong, China; East China Normal University, Shanghai, China
类目: Computer Vision and Pattern Recognition (cs.CV); Computation and Language (cs.CL)
备注:
Abstract:Streaming video understanding requires models to causally update state as video arrives and organize growing history into semantic units that can evolve, persist, and be recalled under bounded computation and memory. This challenge is pronounced in live soccer commentary, where a system must describe completed events, summarize recent play, recall earlier events, or remain silent using only information available before each utterance. We present StreamSoccer, an event-driven system that uses event memory as its intermediate representation. A fixed-budget active memory integrates the stream; completed event states are retained locally and consolidated into retrievable historical records. A unified generator uses current, recent, and historical context to produce three commentary modes, while a rule-assisted scheduler selects a mode or silence. Unlike streaming video-language models organized around frames, visual tokens, or caches, and soccer-commentary methods based on predefined clips or output timestamps, StreamSoccer explicitly models event lifecycles. We construct a three-track streaming soccer commentary dataset and a layered evaluation protocol. At common reference anchors, StreamSoccer obtains CIDEr scores of 38.62, 23.96, and 17.39 for current-event, recent-window, and historical-memory commentary, ranking first on the current-event and historical-memory tracks and second on recent-window. Controlled ablations show that local completed events improve all tracks and that the full system performs best on all three. Across 174 raw-video runs on 58 matches, per-minute RTF p95 ranges from 0.10 to 0.22 without sustained growth with match history. These results indicate that event memory supports streaming soccer commentary across temporal scopes while controlling long-history computation.
[NLP-39] he Asymmetric Harms of LLM Compression
【速读】： 该论文旨在解决大语言模型（Large Language Models, LLMs）压缩过程中隐藏的模型行为异变问题。尽管压缩技术可有效降低部署成本，但传统以困惑度（perplexity）和准确率（accuracy）为代表的聚合指标往往无法揭示压缩带来的深层次影响，如知识保留偏差、模型置信度失真及社会偏见的非对称变化。其解决方案的关键在于通过系统性评估三种主流LLM在11种压缩方法下的表现，从知识保留、模型置信度与社会偏见三个维度进行细粒度分析，发现压缩会显著削弱头部知识（head knowledge）的相对保留率，同时导致模型在丧失新知识后仍对其错误答案保持高置信度；此外，聚合后的偏见评分虽稳定，却可能掩盖不同人口子群体间呈现相反趋势的刻板印象偏好变化。这一发现揭示了压缩引发的非对称行为改变，强调了在模型部署前必须采用精细化评估框架以全面识别潜在风险。
链接: https://arxiv.org/abs/2608.19670
作者: Yuan Wu,Mairui Li,Lesia Semenova,Chudi Zhong
机构: Rice University(莱斯大学); University of North Carolina at Chapel Hill(北卡罗来纳大学教堂山分校); Rutgers University(罗格斯大学)
类目: Computation and Language (cs.CL)
备注:
Abstract:Large language models (LLMs) compression reduces deployment costs, but standard aggregate metrics like perplexity and accuracy often mask underlying behavioral shifts. In this work, we systematically evaluate 3 LLMs across 11 compression methods to investigate the effects of compression on knowledge retention, model confidence, and social bias. We find that compression disproportionately reduces the relative retention of head knowledge compared to tail knowledge. Furthermore, compressed models often remain substantially confident in their incorrect answers on newly lost knowledge. Finally, we demonstrate that stable aggregate bias scores can conceal substantial, opposing shifts in stereotypical preferences across demographic subgroups. Together, these findings reveal asymmetric behavioral changes that aggregate performance measures fail to capture, highlighting the need for granular evaluation of compressed models before deployment.
[NLP-40] ReCache: Efficient KV Cache Reuse and Compression for Tool-Augmented LLM Agents
【速读】： 该论文旨在解决生成式 AI（Generative AI）在代理型语言模型（agentic language models）推理过程中因频繁重复编码跨请求组合与顺序各异的工具与技能模式（tool and skill schemas），导致标准前缀缓存（prefix caching）无法有效复用键值（KV）状态所引发的计算与内存开销过高的问题。其核心解决方案是提出 ReCache 框架，通过资源感知注意力（resource-wise attention）机制，消除跨资源间的交互并分配资源局部位置，生成与组合无关的 KV 块，从而实现对可重用模式的独立缓存；同时，ReCache 采用结构化与语义剪枝策略，仅保留调用关键字段，并通过贡献选择的层-注意力头组路由机制限制资源可见性，以降低推理时的计算与内存消耗。实验结果表明，该框架在保持接近密集调用性能（Inv-F1 82.3% vs. 82.4%）的前提下，实现了 3.655 倍的首次令牌生成速度提升，整体 KV 张量内存占用减少 92.43%，注意力计算加速 1.423 倍，验证了将可重用模式编码与选择性资源访问分离能显著降低代理推理成本，且有效性损失可控。
链接: https://arxiv.org/abs/2608.19662
作者: Yichu Fang,Sitong Wei,Haozhe Hu,Xiaoyu Shen
机构: Shanghai Jiao Tong University (上海交通大学); Eastern Institute of Technology, Ningbo (宁波东方学院)
类目: Computation and Language (cs.CL)
备注: 17 pages, 4 figures
Abstract:Agentic language models repeatedly encode tool and skill schemas that recur across requests in different combinations and orders, preventing standard prefix caching from reusing their key–value (KV) states. We introduce \textbfReCache, a framework for independently caching resource representations while reducing their inference-time computational and memory overhead. Resource-wise attention removes cross-resource interactions and assigns resource-local positions, producing composition-invariant KV blocks. ReCache then restricts resource visibility to contribution-selected layer–KV-head-group routes and retains only invocation-critical fields through structural and semantic pruning. We evaluate ReCache on a benchmark assembled from seven public tool- and skill-use datasets, including resource-disjoint tests. Resource-wise attention matches dense invocation performance (82.3% versus 82.4% Inv-F1) while providing a 3.655 \times time-to-first-token speedup. The complete framework reduces allocated KV-tensor memory by 92.43% and accelerates attention by 1.423 \times . These results show that separating reusable schema encoding from selective resource access substantially reduces agentic inference costs with limited effectiveness loss. The code is available at this https URL.
[NLP-41] Can Agent Memory Systems Track Evolving State?
【速读】： 该论文旨在解决大语言模型（LLM）代理在长期、高风险任务中记忆系统存在的核心缺陷，即现有记忆机制难以有效追踪世界状态的动态演变。传统记忆基准主要关注事实召回类任务，而忽略了在多轮交互过程中事实、约束与决策随时间更新后，模型输出仍需反映最新状态的能力。为此，作者提出“状态追踪”（state tracking）这一关键能力，并构建了StateMemBench基准，涵盖234个跨会话场景，覆盖两种对话长度范式，采用闭池评分机制精确区分答案是否反映当前状态、过时状态或失败，从而将状态追踪错误与其他类型错误解耦。实验表明，现有记忆系统、检索增强基线及长上下文基线均在该任务上表现不佳。针对此问题，作者提出StateMem方法，其核心在于显式建模信息的过时关系（supersession）与语义依赖关系，实现“状态优先”的记忆管理。在DeepSeek-V4-Flash和Qwen-3.5-9B上，StateMem分别相较最强同架构基线提升1.8倍和1.6倍的当前状态准确率。此外，该方法还可作为轻量级单次调用封装集成至现有记忆系统，使六种不同后端的当前状态准确率提升32至67点，其中15至32点可归因于状态结构本身而非额外上下文引入，验证了其高效性与通用性。
链接: https://arxiv.org/abs/2608.19652
作者: Xinyi Fan,Miri Liu,Ruozhen Yang,Siru Ouyang,Jiawei Han
机构: 未知
类目: Artificial Intelligence (cs.AI); Computation and Language (cs.CL)
备注:
Abstract:As LLM-based agents are deployed for longer and higher-stakes tasks, their memory systems continue to have crucial gaps. While existing memory benchmarks focus largely on recall-shaped tasks, we argue an effective memory system must track the evolving state of the world; as facts, constraints, and decisions are revised over a long interaction, answers must reflect the current state and not a superseded one. We define this capability as state tracking and instantiate it in StateMemBench, a benchmark of 234 multi-session scenarios spanning two conversation-length regimes. Its closed-pool grading scores whether an answer reflects the current state, the superseded state, or fails otherwise, separating state-tracking failures from other errors by construction. Our analysis shows that this task is challenging for existing memory systems, retrieval-augmented baselines, and long-context baselines. We then present StateMem, a state-first memory method that explicitly tracks supersession and relational dependencies, and show it improves current-state accuracy over the strongest same-backbone baseline by 1.8x (0.205 - 0.363) on DeepSeek-V4-Flash and over the strongest memory system by 1.6x (0.149 - 0.233) on Qwen-3.5-9B, while remaining competitive with the long-context baselines. Finally, we show the same state approach can be applied as a lightweight single-call wrapper over existing memory systems, lifting current-state accuracy by +32 to +67 points on StateMemBench across six memory and retrieval backends. A length- and cost-matched control attributes +15 to +32 of those points to state structure rather than added context.
[NLP-42] Mitigating Identity Essentialism in LLM Agents with Longitudinal Life Trajectories
【速读】： 该论文旨在解决大语言模型（Large Language Models, LLMs）在社会模拟中因代理（agent）构建方式导致的可信度不足问题，尤其关注现有方法难以真实再现人类个体间多样性与动态变化的缺陷。其核心问题是：静态、孤立的代理表示与仅依赖提示词的记忆机制，导致模型倾向于将群体平均特征误植为个体特质，引发身份本质主义（identity essentialism）倾向，表现为群体内部同质化和人口学维度上的过度分离。解决方案的关键在于提出一种名为LifeMem的纵向记忆框架，通过结合结构化的生命事件检索机制与代理特异的参数化记忆系统，实现对个体经历的持续性整合与动态演化。实验结果表明，LifeMem在Add Health与Understanding Society数据集上显著提升了生成响应在总体分布、组内多样性以及跨生命周期的个体响应变化模式等方面与真实人类数据的一致性，验证了长期生命事件记忆在构建更真实、动态的社会代理中的关键作用。
链接: https://arxiv.org/abs/2608.19621
作者: Hexi Wang,Yujia Zhou,Bangde Du,Weihang Su,Xinyuan Cao,Qingyi Pan,Qingyao Ai,Yueyue Wu,Min Zhang,Yiqun Liu
机构: 未知
类目: Computation and Language (cs.CL)
备注: 23 pages, 12 figures
Abstract:Large language models (LLMs) offer a scalable approach to social simulation, but their credibility depends on how agents are constructed. Existing methods can partially reproduce population-level patterns, yet often fail to capture human-like diversity. Our analysis shows that static-profile agents exhibit stronger demographic separation and within-group compression than humans, a pattern consistent with identity essentialism: demographic labels can encourage models to treat group-average tendencies as individual traits, homogenizing responses within groups. We argue that this limitation arises from two related factors: sparse, static agent representations and the limited ability of prompt-only memory to persistently integrate experience. Inspired by complementary memory systems, we propose LifeMem, a longitudinal memory framework that combines structured life-event retrieval with agent-specific parametric memory for experience integration. Experiments on Add Health and Understanding Society with three LLMs show that LifeMem improves alignment with human data in terms of response distributions, overall and within-group diversity, and patterns of within-person response change across life stages. These findings highlight the value of longitudinal life-event memory for constructing more faithful and dynamically evolving social agents.
[NLP-43] Forking Fast: Efficiently Estimating Uncertainty Dynamics in Text Generation
【速读】： 该论文旨在解决大语言模型（LLM）推理过程中的不确定性分析问题，即如何高效地刻画模型在回答给定问题时可能生成的多种推理路径（reasoning chain）的分布特性。传统基于重采样的分析方法虽能揭示推理过程中关键步骤对最终答案的影响，但其在每个词元（token）或句子层面进行大规模重采样导致计算成本极高。本文的关键解决方案在于提出一种统计建模方法，通过分析大量重采样推理链后发现：尽管单次采样存在噪声，但不确定性动态随样本量增加趋于稳定，表明噪声主要源于采样有限性而非模型对个别推理步骤的敏感性。基于此，作者构建了一种用于平滑低样本量推理轨迹数据的统计模型，使其更接近高样本量的真实分布，从而显著降低采样开销，同时为文本生成中的不确定性动态提供了可解释的统计基础。
链接: https://arxiv.org/abs/2608.19611
作者: Eric Bigelow,Amir Zur,Satchel Grant,Tal Haklay,Can Rager,Owen Lewis,Thomas McGrath,Jack Merullo,Ekdeep Singh Lubana,Atticus Geiger
机构: 未知
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
备注:
Abstract:LLM reasoning is stochastic, and so understanding a model requires grappling with the distribution of reasoning chains that it might produce for a given question, i.e., its uncertainty. Resampling-based analyses characterize this distribution, revealing which steps of a rollout determine how the model arrives at its answer. However, a major limitation of these approaches is that resampling text sequences at every token or sentence in a reasoning chain is very costly. Our work strives to make resampling analysis more computationally efficient, while also shedding light on an important scientific question: what is the right statistical model for explaining uncertainty dynamics in text generation? We show that when resampling many reasoning chains, uncertainty dynamics converge to stable patterns, and noise is largely an artifact of sampling rather than an LLM’s sensitivity to each individual token or reasoning step. We develop a statistical model for smoothing noisy low-sample rollout data to better approximate high-sample data, allowing us to significantly cut sampling costs.
[NLP-44] PEA-DPO: Perception-Enhanced Alignment Direct Preference Optimization for MLLM s Alignment
【速读】： 该论文旨在解决多模态大语言模型（Multimodal Large Language Models, MLLMs）在对齐人类偏好过程中存在的视觉不敏感性（Visual Insensitivity）问题。具体而言，现有基于直接偏好优化（Direct Preference Optimization, DPO）的方法在多模态场景下难以有效识别图像中关键视觉上下文信息，表现为跨图像不敏感（Across-Image Insensitivity）和图像内不敏感（Within-Image Insensitivity）两种形式，导致模型对视觉内容变化缺乏响应能力，进而引发幻觉（hallucination）等问题。其解决方案的关键在于提出感知增强对齐的DPO框架（Perception-Enhanced Alignment DPO, PEA-DPO），通过显式引入视觉偏好信号来强化模型对视觉内容的敏感度，同时在理论上证明该方法可有效缓解上述两类失败模式。实验结果表明，PEA-DPO在多个规模的MLLM上均显著提升了对视觉上下文的感知能力，保持了基础语言建模性能，并有效减少了幻觉现象，实现了更强的多模态对齐效果。
链接: https://arxiv.org/abs/2608.19598
作者: Jiawei Feng,Jiancan Wu,Xingyu Zhu,Junkang Wu,Xiang Wang,Xiangnan He
机构: University of Science and Technology of China(中国科学技术大学); National University of Singapore(新加坡国立大学)
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI); Computation and Language (cs.CL); Multimedia (cs.MM)
备注:
Abstract:Direct Preference Optimization (DPO) has emerged as an effective approach for aligning large language models (LLMs) with human preferences. However, its adaptation to multimodal settings remains unexplored. Through representational analysis, we identify a key limitation in multimodal preference optimization, which we term visual insensitivity: models often fail to distinguish between images and those with critical visual context removed. Our theoretical analysis further uncovers two manifestations of this problem, namely Across-Image Insensitivity and Within-Image Insensitivity. To address these challenges, we propose Perception-Enhanced Alignment DPO (PEA-DPO), a framework for multimodal LLMs alignment, which explicitly leverages visual preference signals to overcome visual insensitivity. We further provide a theoretical analysis demonstrating that PEA-DPO provably mitigates both failure modes. Empirical results demonstrate that PEA-DPO enhances sensitivity to visual context while preserving the language modeling capacity of the base model. Evaluations across three hallucination benchmarks using MLLMs of varying scales show that PEA-DPO effectively mitigates visual insensitivity, achieves stronger multimodal alignment, and substantially reduces hallucinations.
[NLP-45] Remember Verify or Ask? Cross-Family Evaluation of Memory Commitment in LLM Agents
【速读】： 该论文旨在解决大语言模型（Large Language Model, LLM）在长期交互中因持久化记忆（persistent memory）更新错误而导致行为隐性偏差的问题。核心挑战在于界定“记忆澄清边界”（memory-clarification boundary），即如何决策：对于交互中获取的信息，应将其持久化存储、仅在当前上下文中使用、重新验证，还是向用户澄清确认。为系统评估这一边界，研究构建了MCB数据集，包含140个主场景（70个开发集与70个保留测试集）及70个对照集，涵盖动作标签与结构化工具调用选择的双重评估维度。研究表明，尽管模型在验证事实变更方面表现优于主动请求用户澄清，但其澄清召回率仍极低（如裸模型在12个需澄清项中未提出任何询问）。通过少样本提示（few-shot prompting）可显著提升准确率（从0.557升至0.771，p_H = 0.002），但澄清召回率仅达0.333；而策略性提示（policy prompt）虽有效降低错误持久化率（从0.243降至0.100，p_H = 0.038），但对整体准确率提升不显著。此外，模型与人工标注间在工具调用选择上的一致性较低（Claude模型约57%，Qwen仅为23%），且Qwen在引入提示后准确率下降至0.343（p_H = 0.047），凸显模型在真实任务中执行记忆管理能力的严重不足。因此，解决方案的关键在于设计能够同时引导模型做出合理决策和正确工具调用的多维度评估框架与提示机制，以确保记忆更新的可靠性与可解释性。
链接: https://arxiv.org/abs/2608.19564
作者: Baichuan Li,Junyi Yao,Zihao Zheng
机构: Southern Methodist University (南卫理公会大学); Washington University in St. Louis (圣路易斯华盛顿大学)
类目: Computation and Language (cs.CL)
备注:
Abstract:Persistent memory can personalize an LLM agent, but an incorrect durable update can silently distort future behavior. We study the memory-clarification boundary: whether interaction-derived information should be persisted, used only in the current context, re-verified, or clarified with the user. MCB contains 140 primary scenarios, split into 70 development and 70 held-out items, plus a separate 70-item contrast set. It evaluates both action labels and structured tool-call selection. Two non-authors independently label the 70 held-out primary and 70 contrast items (97.1% agreement, Cohen’s kappa = 0.962); a blind third resolves four disagreements, replacing eight author labels by non-author majority. Across Claude and Qwen, models verify changing facts more reliably than they ask users to resolve ambiguity. Bare Qwen asks on 0/12 clarification items while verifying 12/18 freshness items. Few-shot prompting raises accuracy from 0.557 to 0.771 (paired delta = +0.214, Holm-adjusted exact McNemar p_H = 0.002), yet clarification recall remains 0.333. The policy prompt reduces erroneous persistence from 0.243 to 0.100 (p_H = 0.038), although its accuracy gain is not significant. Label-tool agreement is 57% for each Claude model and 23% for Qwen; Qwen accuracy falls from 0.557 to 0.343 (p_H = 0.047). Memory evaluation must test both stated decisions and tool-call choices.
[NLP-46] Reliable Financial Named Entity Recognition under Domain Shift
【速读】： 该论文旨在解决金融领域生成式人工智能（Generative AI）系统在跨文本类型部署时面临的可靠性问题，即标准的F1分数无法有效反映模型在输入分布发生变化时哪些预测仍可安全自动化。针对金融命名实体识别（NER）任务，研究构建了一个三层次压力测试框架，涵盖美国证券交易委员会（SEC）文件、财经新闻及通用主题社交媒体内容，以模拟极端域外情况。其解决方案的关键在于通过五种推理阶段置信度信号（包括整体输出概率、实体跨度概率与自一致性等）评估模型在分布偏移下的表现，并提出基于置信度的择优预测（selective prediction）策略。研究发现，整体输出概率虽在域内表现优异，但在域外性能显著下降；而实体跨度概率和自一致性更具鲁棒性，其中自一致性不仅更优且无需后处理校准。尽管弃权机制可在域内将句子错误率从34.3%降至2%以下，并在财经新闻中保持有效性，但在极端社交媒体分布偏移下无法恢复出可用的高质量子集。因此，研究提出一种分阶段部署策略：先在上游检测严重分布偏移，再实施预测级别的置信度过滤，从而提升金融AI系统的安全性与可信度。
链接: https://arxiv.org/abs/2608.19558
作者: Zihao Zheng,Baichuan Li,Junyi Yao,Jiayu Long
机构: Washington University in St. Louis (圣路易斯华盛顿大学); Southern Methodist University (南卫理公会大学)
类目: Computation and Language (cs.CL)
备注:
Abstract:Financial AI systems often train information extractors on one textual register and deploy them across filings, news, and user-generated content, while standard F1 scores do not indicate which predictions remain safe to automate when the input distribution changes. We study confidence estimation and selective prediction for financial named entity recognition (NER) on a three-tier stress test spanning SEC filings, financial news, and general-topic social media as an extreme out-of-domain condition. We evaluate a BERT tagger and LoRA-tuned Qwen2.5-0.5B/1.5B models using five inference-time confidence signals, three training seeds, and bootstrap intervals. Confidence rankings themselves change under distribution shift: whole-output probability is the strongest in-domain error detector but deteriorates out of domain, whereas entity-span probability and self-consistency are more robust; self-consistency is also better calibrated without post-hoc fitting. Abstention reduces sentence error from 34.3% to below 2% on the highest-confidence 40% of in-domain inputs and remains useful on financial news, but recovers no usefully large clean subset under the extreme social-media shift. These results motivate a staged deployment strategy that detects severe distribution shift upstream before applying prediction-level confidence gating.
[NLP-47] When Machines Speak: A Unified Generative Framework for Integrating Machine-Native Symbols into Pretrained Large Language Models
【速读】： 该论文旨在解决预训练大语言模型（LLM）在处理机器原生符号（machine-native symbols）时存在的根本性割裂问题：尽管现实世界中的许多AI系统使用离散的、结构化的机器原生符号来表示实体、行为和信息，但这些符号并不属于自然语言的词元空间，导致语言建模与结构化预测之间存在鸿沟。现有方法通常需要将符号转化为自然语言或设计特定任务架构，增加了复杂性且限制了泛化能力。论文提出的解决方案——UniLang，其关键在于通过扩展预训练大语言模型的词汇表和嵌入空间，将机器原生符号作为与自然语言词元同等地位的生成单元进行建模，实现文本与符号的联合自回归生成。这一统一接口使预训练大语言模型能够直接操作机器原生表示，无需显式语义化或依赖特定任务结构。在序列推荐和法律先例预测两个结构差异显著的任务上，UniLang均显著优于强基线模型，验证了其作为异构机器原生表示通用生成建模骨干的潜力。
链接: https://arxiv.org/abs/2608.19529
作者: Su Yan,Rakesh Iyer
机构: 未知
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
备注: Code: this https URL
Abstract:Many real-world AI systems represent entities, behaviors, and structured information using discrete machine-native symbols rather than natural language. While these representations are compact and preserve task-relevant structure, they lie outside the linguistic token space of pretrained large language models (LLMs), creating a fundamental divide between language modeling and structured prediction. We introduce UniLang, a unified generative framework that bridges this divide by extending pretrained LLMs to treat machine-native symbols as first-class generative units alongside natural-language tokens. UniLang expands the LLM’s vocabulary and embedding space with grounded machine-native representations, enabling textual and symbolic tokens to be jointly modeled and generated under a single autoregressive objective. This unified interface allows pretrained LLMs to directly operate on machine-native representations without requiring them to be verbalized as natural language or relying on task-specific architectures. We evaluate UniLang on two structurally distinct tasks, sequential recommendation and legal precedent prediction, spanning different domains and types of structured prediction. Across both tasks, UniLang consistently outperforms strong baselines, demonstrating a path toward extending pretrained LLMs beyond language and using them as a common generative modeling backbone for heterogeneous machine-native representations.
[NLP-48] Hear2Act: Benchmarking When Prosody Should Change What an Assistant Does
【速读】： 该论文旨在解决现有评估基准在任务导向对话中孤立评价韵律感知、响应适当性与对话能力的问题，难以检验韵律线索是否真正影响下游决策。其核心挑战在于：当语义内容不变时，仅通过韵律（Prosody）传递的关键信息能否有效引导智能体做出正确判断。为此，作者提出Hear2Act——一个统一的评估协议，涵盖480个基于人物设定的情境，包含隐藏用户关切和可客观验证的结果。关键创新在于控制变量设计：保持任务与用户需求一致，仅改变关切是通过显性词汇表达还是主要依赖韵律传达，并在文本、音频及关切状态三种条件下评估模型决策。实验结果表明，在缺乏显性词汇线索时，音频信息虽能被模型感知并转化为文本表示，但若无明确的中间表征环节，其对后续动作选择的影响极为有限；而一旦引入对语音中隐含关切的状态推断与文本化表达，决策准确率显著提升至39.6%，接近使用真实状态时的40.7%。这揭示了音频能力大语言模型（audio-capable LLMs）在处理隐性韵律信息时的关键瓶颈：必须通过显式中间表征将语音中的隐含语用信息转化为可操作的文本状态，才能实现有效的行动转化。
链接: https://arxiv.org/abs/2608.19515
作者: Xinyi Liu,Hooshang Nayyeri,Dilek Hakkani-Tur,Emine Yilmaz,JK Kim,Yifei Zhang,Charith Peris,Hari Thadakamalla
机构: Amazon(亚马逊); University of Illinois Urbana-Champaign (伊利诺伊大学香槟分校); University College London (伦敦大学学院)
类目: Computation and Language (cs.CL)
备注:
Abstract:Prosodic cues can convey task-relevant information that alters the trajectory and outcome of a task-oriented dialogue, even when the words themselves remain unchanged. Yet existing benchmarks typically evaluate prosodic perception, response appropriateness, and task-oriented dialogue in isolation, making it difficult to test whether prosodic evidence changes downstream decisions. We introduce Hear2Act, a unified evaluation protocol for text and spoken assistants with 480 persona-grounded scenarios, hidden user concerns, and objectively verifiable outcomes. For each scenario, we keep the task and user needs fixed while varying whether the same concern is conveyed explicitly in words or primarily through prosody, and evaluate decisions under transcript, audio, and concern-state access. Using Hear2Act, we evaluate two audio-capable LLMs. Under Prosody-mediated feedback, adding audio to the transcript changes the average optimal-solution rate only from 14.6% to 15.3%. In contrast, when models infer the concern status from audio, represent it in text, and use it for next-action selection, the rate rises to 39.6%, close to 40.7% with the ground-truth state. This contrast, however, largely disappears under Explicit lexical feedback, where the concern is verbally mentioned in the utterance. Together, these results show that prosody matters when lexical evidence is insufficient, and that audio-capable LLMs can recover information from speech but do not reliably carry it into action without an explicit intermediate representation. Subjects: Computation and Language (cs.CL) Cite as: arXiv:2608.19515 [cs.CL] (or arXiv:2608.19515v1 [cs.CL] for this version) https://doi.org/10.48550/arXiv.2608.19515 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
[NLP-49] DeltaMomentum: A Key-Value based Anisotropic Momentum Update via Delta Rule
【速读】： 该论文旨在解决现代优化器在处理深度神经网络训练过程中输入数据高度各向异性（anisotropic）时的效率瓶颈问题。传统优化器采用指数移动平均（Exponential Moving Average, EMA）形式的动量，对所有梯度方向以固定速率遗忘，无法适应不同方向出现频率的差异，导致对高频方向响应过度而对低频方向更新不足。其解决方案的关键在于提出DeltaMomentum，一种将方向感知能力内嵌于动量更新规则的新机制。核心思想是将线性层的梯度分解为“输入侧作为键（key）”与“输出侧误差作为值（value）”的结构，并基于此采用经典的**增量规则（delta rule）**进行动量缓冲区更新，使得每个方向的遗忘速率由其出现频率动态决定。该方法不仅被证明是有效的动量形式，且无需矩阵求逆即可实现输入侧曲率校正，同时在固定和漂移最优解场景下均能更快速清除陈旧方向。此外，DeltaMomentum可作为任意优化器的动量缓冲区直接替换，其系数在模型宽度变化下具备良好迁移性，额外计算开销仅为门控MLP块线性计算成本的22.2%至25.0%，且无持久内存占用。实验表明，在FineWeb-Edu预训练中，采用DeltaMomentum的AdamW（DeltaAdamW）相比原版AdamW在67M和370M参数规模下分别减少46.39±4.32%和22.12±0.80%的训练步数达到相同验证损失，1B规模下仍保持优势；在多种任务（如SGD、ResNet-18、ViT-Tiny on CIFAR-10）上亦表现更优，训练诊断进一步验证了其更优的梯度追踪能力和更健康的输入方向状态。
链接: https://arxiv.org/abs/2608.19491
作者: Euijin Hong,Guannan Qu
机构: Carnegie Mellon University (卡内基梅隆大学)
类目: Machine Learning (cs.LG); Computation and Language (cs.CL); Optimization and Control (math.OC); Machine Learning (stat.ML)
备注:
Abstract:Most modern optimizers form their momentum as an exponential moving average (EMA) of past gradients, forgetting every direction at one fixed rate. However, the inputs a deep network sees during training can be highly anisotropic, with a few directions queried frequently while most are seen rarely. Recent methods address this anisotropy by wrapping extra processing around this buffer, leaving the momentum update itself unchanged. We propose DeltaMomentum, which builds direction-awareness into the momentum update rule. The main observation is that the gradient of a linear layer splits into an input that acts as a key and an output-side error that acts as a value. Exploiting the key-value structure, DeltaMomentum updates the momentum buffer by the canonical delta rule, so each direction is forgotten at a rate set by how often it appears. We prove that it is a valid momentum, that it applies the input-side curvature correction without matrix inversion, and that it clears stale directions faster than EMA under both a fixed and a drifting optimum. It is a drop-in replacement for the momentum buffer of any optimizer, its coefficient transfers across widths under \mu P, and its extra compute stays between 22.2% and 25.0% of a gated-MLP block’s linear cost with no persistent memory. In FineWeb-Edu pretraining, AdamW with DeltaMomentum (DeltaAdamW) reaches AdamW’s validation loss in up to 46.39 \pm 4.32% fewer steps at 67M and 22.12 \pm 0.80% at 370M over three seeds, and the gain persists at 1B on a Chinchilla-optimal budget. A Muon baseline tuned under the same protocol sits above DeltaAdamW at both language-model scales, and the gain holds for SGD, ResNet-18, and ViT-Tiny on CIFAR-10. Training-time diagnostics confirm the predicted mechanism, better gradient tracking and healthier input directions.
[NLP-50] Measuring What a Specification Determines: A Formal Semantic-Block Model and an Execution-Judged Benchmark
【速读】： 该论文旨在解决大语言模型（Large Language Models, LLMs）在生成系统规格说明时存在的质量评估难题，尤其是如何在不依赖模型能力的前提下，客观衡量规格说明本身的清晰性与一致性。其核心问题在于：现有评估方法往往将模型性能与规格质量混同，导致无法准确判断规格本身是否具备可实现性和确定性。为此，论文提出了一种形式化的语义块（semantic-block）模型，将规格说明建模为包含语义块、依赖关系、块内规则、决策点及显式开放问题的结构化体系，并施加四项机器可验证的良构性条件（无环性、单一所有权、约束支配性与完备性或歧义终止），以确保规格的逻辑严谨性。解决方案的关键在于引入“执行判断基准”（execution-judged benchmark），通过固定实施者小组、设置无规格对照组，并利用真实运行的PostgreSQL 16和活体Oracle实例作为确定性执行裁判，实现对规格质量的独立评估。研究结果表明，该语义块模型在迁移任务中实现了约71%的上下文压缩率，覆盖85.5%的预定义构造分类体系，且在多种测试中表现出高度可恢复性；然而，尽管“确定性”（determinacy）作为形式概念得到支持，其作为独立经验质量指标在当前主流LLM实现者中的适用性仍受限，揭示了规格质量评估需超越单纯收敛性指标的必要性。
链接: https://arxiv.org/abs/2608.19475
作者: Oleg Grynets,Dmytro Kostetskyi,Vasyl Lyashkevych
机构: 未知
类目: oftware Engineering (cs.SE); Artificial Intelligence (cs.AI); Computation and Language (cs.CL); Logic in Computer Science (cs.LO)
备注: 11 pages, 1 figure, 7 tables, 37 references
Abstract:This work introduces a formal semantic-block model for specifications and an execution-judged benchmark for evaluating specification quality independently of model capability. A specification is represented as a structure comprising semantic blocks, dependency relations, block-owned rules, decision points, and explicitly open questions, subject to four machine-checkable well-formedness conditions: acyclicity, single ownership, constraint domination, and totality or ambiguity-stop. Determinacy is defined model-theoretically as agreement among all conforming implementations and is estimated empirically through convergence across independent implementers. The model is instantiated on an Oracle-to-PostgreSQL migration specification containing 18 blocks and 19 dependency edges. Computational validation shows that the five-layer decomposition reduces mean per-task context by approximately 71% through dependency closures, covers 85.5% of the study-defined Oracle construct taxonomy with all identified gaps triaged, is not Pareto-dominated by the tested alternative partitions, and is recovered at the 99.9th percentile from citation-derived edges not used to define the original structure. The benchmark keeps the implementer panel fixed, includes a mandatory no-specification control arm, and uses PostgreSQL 16 and a live Oracle instance as deterministic execution judges. Six designed studies, including three pre-registered manipulations and three diagnostic analyses, further examine specification effects. Repeated runs on a 25-unit subsample reveal an empirical variability floor with a median arm-delta spread of 14.4 percentage points. The results support determinacy as a formal concept but not as a standalone empirical quality metric for the evaluated contemporary LLM implementers.
[NLP-51] SynFlow: A Multidimensional Diachronic Semantic Analysis Toolkit
【速读】： 该论文旨在解决传统词汇语义变迁（Lexical Semantic Change, LSC）建模方法在解释性方面的局限性问题，即向量空间模型虽能捕捉语义变化，却难以揭示具体哪些语言使用维度（如句法行为、形态特征、构式模式等）发生了演变。现有历时语料库研究虽能分析可解释的语言维度，但通常采用独立的分析流程，缺乏整合性。为此，本文提出SynFlow——一个开源的多维历时语言使用分析工具包，其核心创新在于将多种语言学观察（包括基于依存关系的共现、形态特征、构式配置及外部生成的框架语义表示）统一转化为时期特定的分布，并通过共享分析工作流实现跨维度整合。关键解决方案包括：支持多种距离度量、逐值分解、统计检验以及词项的增量聚类，从而在多个语言维度上同步追踪语义变迁。通过德语形容词“viral”的案例研究，展示了单一语义发展如何在句法、词汇、构式和形态等多个层面体现；同时结合SemEval-2020 Task 1的已有结果，验证了该方法在语义变迁检测任务中的有效性与竞争力。
链接: https://arxiv.org/abs/2608.19472
作者: Bach Phan-Tat,Kris Heylen,Dirk Geeraerts,Stefano De Pascale,Dirk Speelman
机构: KU Leuven(鲁汶大学); Instituut voor de Nederlandse Taal(荷兰语研究所); Vrije Universiteit Brussel(布鲁塞尔自由大学)
类目: Computation and Language (cs.CL)
备注:
Abstract:Lexical semantic change (LSC) is commonly modelled through vector-space representations, but these approaches often provide limited insight into which aspects of usage are changing. Diachronic corpus research instead examines interpretable dimensions such as syntactic behaviour, morphology, and constructional patterns, but typically through separate analytical workflows. We present SynFlow, an open-source toolkit for multidimensional diachronic analysis of linguistic usage. SynFlow converts linguistic observations into period-specific distributions and applies a shared workflow across dependency-based co-occurrences, morphological features, constructional configurations, and externally derived representations such as Frame Semantics. It supports different distance measures, together with value-level decomposition, statistical testing, and incremental clustering of lexical fillers. We demonstrate SynFlow through a qualitative case study of the German adjective viral, showing how a single semantic development is reflected across syntactic, lexical, constructional, and morphological dimensions. We further report previously published results on SemEval-2020 Task 1 to situate the performance of these representations relative to existing lexical semantic change detection systems.
[NLP-52] Are LLM s becoming similarly creative? Evidence from three years of models
【速读】： 该论文旨在解决大语言模型（Large Language Model, LLM）在开放性任务中表现演变趋势不明确的问题，尤其关注创造性、原创性和多样性等非可验证性指标的动态变化。传统评估多聚焦于答案可验证的任务，而忽视了对开放式生成任务中模型创造力的发展评估。本文的关键解决方案是基于句子嵌入相似度（sentence-embedding similarity），对三年间多个模型版本在真实用户查询集合Infinity-Chat100与经典心理学创造力测评工具Alternate Uses Task上的输出进行分析。研究发现，尽管模型整体生成质量可能提升，但其输出多样性呈现显著下降趋势，表明不同模型的创造性内容趋于同质化。这一现象若持续发展，可能导致人机协同创作过程中人类主体性的弱化，因而需重新审视大语言模型在人类创造性活动中的角色定位。
链接: https://arxiv.org/abs/2608.19437
作者: Nirav Patel,Josiah Crossman,Eva Aggarwal,Emily Wenger
机构: Duke University (杜克大学); Duke University (杜克大学)
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI); Computers and Society (cs.CY)
备注: 12 pages, 4 figures
Abstract:Many benchmarks track Large Language Model (LLM) performance on tasks with verifiable answers, but less is known about how LLM performance is evolving on open-ended tasks, where creativity, originality and diversity may matter as much as quality. As LLMs increasingly support human ideation and creative work, understanding trends in LLM performance on open-ended tasks is critical. This paper presents a preliminary analysis of LLM creative outputs spanning three years of model releases, examining model responses to Infinity-Chat100, a real-world collection of open-ended user queries, and the Alternate Uses Task, an established psychometric creativity assessment. Using sentence-embedding similarity, we examine trends in LLM responses to these prompts. Our findings show a statistically significant decrease in model output diversity over time, suggesting that LLM outputs may be converging in creative substance across models. If this trend persists, LLM-driven homogenization may progressively diminish human agency in human-AI co-creative work, demanding careful consideration of LLMs’ role in the human creative process.
[NLP-53] Linguistic Holonomy and Statistical Watermarks: Inner Geometry of Meaning-Preserving Transformations
【速读】： 该论文旨在解决生成式语言模型中统计水印（statistical watermarks）在语义保持变换下的脆弱性问题，即水印信号易被形式变化但语义不变的文本改写所削弱甚至消除。其核心挑战在于现有评估方法仅依赖原始文本与重写文本之间的语义相似度作为衡量标准，而这一“终点统计量”无法捕捉变换路径中的内在结构信息。本文的关键解决方案是引入语言环（linguistic loops）的数学框架，证明语义保持变换链的不变量可分解为两部分：末端项（endpoint）与位于初始状态稳定子群内的全息项（holonomy）。其中，全息项对应于嵌入空间单位球面上的平行移动（parallel transport），使类比于规范场论中的威尔逊环（Wilson loop）从修辞上升为严格定理。在检测端，作者建立了精确恒等式：残留统计量正比于未被破坏的种子窗口位置数，由此导出独立编辑情形下的衰减律 ρ^h+1。该恒等式揭示了一个令人震惊的后果——在相同保留率下，残余信号可能为原始值的1/2、1/4或完全消失，具体取决于编辑位置分布，这表明传统基于终点相似度的评估具有根本性缺陷。
链接: https://arxiv.org/abs/2608.19369
作者: Daniele Corradetti
机构: Grupo de Física Matemática, Instituto Superior Técnico (数学物理组，里斯本理工学院); Departamento de Matemática, Universidade do Algarve (阿尔加维大学数学系)
类目: Computation and Language (cs.CL); Cryptography and Security (cs.CR); Differential Geometry (math.DG)
备注:
Abstract:Statistical watermarks for language models live in the freedom of the signifier: they choose among tokens that are nearly equivalent in meaning, and they are therefore eroded by exactly those transformations which move the form of a text while leaving its content in place. The literature measures such transformations by their endpoint, through the semantic similarity between the original and the rewritten text. We show that the endpoint is the wrong statistic. Adapting the formalism of linguistic loops, we prove that the invariant of a chain of meaning-preserving transformations factorises canonically into an endpoint part and a holonomy in the stabiliser of the initial state, the second of which the semantic deficit cannot see; the loop rotation is parallel transport on the unit sphere of the embedding space, so that the analogy with the Wilson loop becomes a theorem rather than a figure of speech. On the side of the detector we prove an exact identity: the residual statistic is proportional to the number of positions whose seeding window survived intact, from which the decay law \rho^h+1 follows as the independent-edit corollary. The identity has a disconcerting consequence, which we confirm to three decimal places: at one and the same retention rate the surviving signal may be one half of the original, one quarter of it, or exactly nothing, according only to where the edits fall.
[NLP-54] A Speech Corpus for Mizo Automatic Speech Recognition: Whisper and SraVaani 1.0 Fine-Tuning with Morphology-Aware Evaluation
【速读】： 该论文旨在解决低资源语言米佐语（Mizo）的自动语音识别（ASR）系统构建问题，尤其针对其语音数据稀缺、缺乏成熟标注资源的挑战。解决方案的关键在于通过收集并精心整理17.62小时的米佐语语音数据，结合多语言预训练模型进行微调：分别采用三种Whisper多语言模型以及SraVaani 1.0印地语多语言模型进行实验。研究发现，Whisper-large-v3在常规词错误率（WER）上表现最佳（18.08%），而引入形态感知评估后进一步降至7.22%；尽管SraVaani 1.0在零样本条件下表现较差（WER 58.27%），但经过针对米佐语的精细化数据微调后，其性能显著提升，常规WER降至29.45%，形态感知WER降至17.93%。这表明，高质量的数据标注与针对性微调是提升低资源语言ASR性能的核心因素，且预训练模型在适配新语言时具备较强的泛化能力。
链接: https://arxiv.org/abs/2608.19361
作者: Priyankoo Sarmah,Sanasam Ranbir Singh,Lalhmingmawia
机构: 未知
类目: Computation and Language (cs.CL); Audio and Speech Processing (eess.AS)
备注:
Abstract:This study reports the development of an Automatic Speech Recognition (ASR) system in Mizo, a low-resource language. The development included collecting 17.62 hours of speech data, curating it, and fine-tuning the Mizo ASR system with three Whisper multilingual models and with the SraVaani 1.0 Indic multilingual model. Whisper-large-v3 achieved the lowest conventional WER (18.08%), while morphology-aware evaluation yielded a WER of 7.22%. Zero-shot evaluation of the SraVaani 1.0 Indic multilingual model yielded a WER of 58.27%, while Mizo-specific fine-tuning reduced the conventional WER to 29.45% and the morphology-aware WER to 17.93%. The results demonstrate that the Whisper model can achieve a substantially low WER, even when adapted to an unseen language. In contrast, SraVaani 1.0 supports the Mizo language in its multilingual model; however, fine-tuning with carefully curated Mizo speech data substantially improves its performance.
[NLP-55] Outcome Monitors: Recovery Affordances for Silent Tool Failures
【速读】： 该论文旨在解决在复杂任务执行过程中，当工具调用（tool call）因超时或失败导致结果不可用时，智能体（agent）缺乏有效机制识别并应对异常输出的问题。现有系统往往将错误响应（如缓存的错误页面或负价格等非预期结果）误认为有效事实，从而引发任务中断或错误推理。其解决方案的关键在于提出结果监控器（Outcome Monitors），该机制通过从任务无关的执行轨迹中挖掘或基于公开模式（public schemas）推导出“结果契约”（outcome contracts），实时检测实际输出是否违反这些契约。一旦检测到违规，监控器会保留原始结果，并生成一份非约束性收据（nonbinding receipt），明确指出被违反的属性及可调用的公共恢复工具（recovery tools）。实验表明，在引入故障注入的冻结评估中，该方法使ToolMaze任务完成率从10.9%提升至28.1%，在两个供应商系列的四个模型上均实现显著增益，并在第三个供应商体系中复现；在tau-bench零售基准测试中，任务完成度分别提升14.0和12.0个百分点。控制实验进一步验证了恢复工具列表是性能提升的核心因素，移除后收益消失，恢复后效果重现，而诊断信息与时间戳则无显著影响。此外，收益主要集中在故障导致任务无法继续的场景；当检测范围超出已挖掘词汇表时，检测准确率下降至46%，但任务仍能持续执行且完成率未变，表明当前方法对合同词汇外的异常检测能力有限，拓展检测边界仍是开放问题。
链接: https://arxiv.org/abs/2608.19303
作者: Sugam Panthi,Rabab Abdelfattah
机构: 未知
类目: Artificial Intelligence (cs.AI); Computation and Language (cs.CL); Software Engineering (cs.SE)
备注: 16 pages (9 main + 7 pages supplementary material), 3 figures
Abstract:When a tool call times out, the agent sees the failure and can route around it. A cached error page or negative price can instead arrive in the expected format and be consumed as fact. We introduce Outcome Monitors, which detect violations of outcome contracts mined from task-disjoint traces or derived from public schemas. On a violation, the monitor preserves the result and issues a nonbinding receipt naming the violated property and public recovery tools. In frozen, prespecified evaluations with injected failures, Outcome Monitors raise ToolMaze completion from 10.9% to 28.1% across four models in two provider families and replicate in a third. In tau-bench retail, completion improves by 14.0 and 12.0 points on two tiers. In separate ToolMaze controls, removing the recovery-tool list eliminates the measured gain and restoring it recovers the effect; diagnostic detail and timing produce no detectable differences. Gains concentrate where the fault blocks completion. On a suite transcribed from a published incident taxonomy, detection outside the mined vocabulary falls to 46%, though delivery continues and completion is unchanged. Recovery tools are the active receipt content in these controls; extending detection beyond the contract vocabulary remains open.
[NLP-56] Can Conversational AI loosen Us-Versus-Them Boundaries? The Effects of Common Dual and Separate Identity Framings on Pro-Immigrant Intergroup Helping
【速读】： 该论文旨在解决因移民人数上升而加剧的群体间紧张关系问题，尤其关注主流群体成员对拉美裔移民（Latine immigrants）的刻板认知与社会疏离。传统偏见减少项目面临可扩展性差和美国政策限制等挑战，因此亟需新型干预手段。其核心解决方案是利用生成式 AI（Generative AI）通过对话引导非拉美裔白人美国成年人重构对拉美裔移民的身份认知。研究基于共属内群体认同模型（Common Ingroup Identity Model），设计五轮与大语言模型（GPT-4o）的对话实验，将模型引导至三种身份框架：共享的超群体身份（共同美国人身份）、双重身份（既是拉美裔又是美国人）或分离身份（强调文化边界差异），并设置无关话题作为对照。结果显示，强调共属身份（共同或双重身份）显著降低了“我们 vs. 他们”的二元分类，提升了双重身份认同；尽管对亲多样性信念和行为的直接影响不显著，但两类共属身份条件均显著提升了行动意愿。路径分析表明，这种效应主要通过降低分离性分类产生间接影响，且语义相似性分析证实对话内容忠实遵循预设叙事——参与者对共享身份语言的趋同程度越高，行动意愿越强；对分离性语言的趋同则与行动意愿呈负相关。这些发现表明，短暂的生成式人工智能对话可在认知层面松动群体边界，但认知重构与实际行为之间仍存在显著差距，提示未来干预需兼顾认知转变与行为激励机制。
链接: https://arxiv.org/abs/2608.19220
作者: Oluwadamilola Jeboda,John F. Dovidio,Jonas R. Kunst
机构: 未知
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
备注:
Abstract:Rising immigration has intensified intergroup tensions in many countries. Traditional bias-reduction programs remain difficult to scale and increasingly constrained by U.S. policy. This preregistered experiment tested whether conversational AI can shift how majority-group members categorize and relate to Latine immigrants. Drawing on the common ingroup identity model, a quota-representative national sample of 658 non-Latine White U.S. adults completed five rounds of dialogue with a LLM (GPT-4o). The model was instructed to frame Latine immigrants in terms of a common ingroup identity (a shared American identity), a dual identity (both Latine and American), or a separate identity (distinct cultural boundaries), or to discuss an unrelated topic in a control condition. The manipulations altered categorization: relative to control, common ingroup identity and dual identity conversations lowered separate categorization, and dual identity conversations raised dual categorization. Although direct effects on behavior and pro-diversity beliefs were nonsignificant, willingness to act was significantly higher in the conditions emphasizing a superordinate identity (common ingroup and dual identity). A path model further revealed indirect associations: both conditions reduced separate categorization, which in turn correlated with greater willingness to act. Semantic similarity analyses of the transcripts confirmed that conversations tracked their assigned narratives; participants’ convergence with shared-identity language related positively, and with separate-identity language negatively, to willingness to act. These effects were largely consistent across moderators (need for closure, openness to experience, and political orientation). The findings show that brief AI conversations can loosen us-versus-them boundaries while underscoring the gap between cognitive recategorization and behavior.
[NLP-57] me-Series Retrieval for Grounding Multimodal Language Models in Remaining Useful Life
【速读】： 该论文旨在解决在实际故障预测与健康管理（PHM）场景中，如何提升生成式人工智能（Generative AI）系统在剩余使用寿命（RUL）估计任务中的性能与可靠性问题。其核心挑战在于，尽管多模态大语言模型（MLLM）具备强大的语义理解能力，但在缺乏有效上下文支持的情况下，其对时间序列退化轨迹的推理能力受限。为此，论文提出一种基于时间序列检索增强生成（Time-series Retrieval-Augmented Generation, RAG）的框架：通过从训练集中检索历史相似的退化段落，并将其与待测轨迹共同构建为结构化的视觉对比表征，输入至MLLM进行多模态推理。该方案的关键在于利用可解释的、基于实例的检索机制，为MLLM提供更具针对性的上下文证据，从而增强其对复杂退化模式的理解能力。实验结果表明，相较于随机参考选择的基线方法，该检索策略显著降低了RUL预测误差并提升了模型稳定性，且其增益程度与模型容量正相关，说明高容量MLLM更善于利用检索到的证据进行精准推断。研究证实，时间序列RAG是提升多模态故障预测推理能力的有效途径，同时也揭示了当前基于MLLM的RUL估计在真实工业环境下的局限性。
链接: https://arxiv.org/abs/2608.19218
作者: Valeriu Dimidov,Raphaël Frank
机构: Interdisciplinary Centre for Security, Reliability and Trust (SnT) University of Luxembourg (卢森堡大学安全、可靠性和信任跨学科中心)
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
备注:
Abstract:Large language models (LLMs) and agentic AI systems are increasingly being explored for domain-specific maintenance and prognostics tasks, raising the question of whether they can effectively support prognostics and health management (PHM). In this paper, we investigate remaining useful life (RUL) estimation with multimodal large language models (MLLMs) grounded through time-series retrieval. We propose a framework in which historically similar degradation segments are retrieved from the training set and, together with the test trajectory, transformed into a visual comparison artifact that is processed by the MLLM through a structured multimodal prompt. The approach is evaluated on the FD001 partition of the C-MAPSS benchmark under repeated experiments comparing retrieval-based inference against a non-retrieval baseline based on random reference selection. The results show that time-series retrieval consistently improves MLLM-based RUL prediction across the evaluated models, yielding lower error and more stable performance. At the same time, the magnitude of the benefit depends on model capacity, indicating that retrieval is most effective when the underlying MLLM is able to exploit the retrieved evidence. Overall, the study shows that time-series RAG is a promising mechanism for improving multimodal prognostic reasoning, while also highlighting the current limitations of MLLM-based RUL estimation in practical PHM settings.
[NLP-58] NepOOC-M: Bilingual Nepali-English Benchmark and Comparative Analysis of Multimodal Architectures for OOC Detection
【速读】： 该论文旨在解决尼泊尔语环境中“上下文无关”（Out-of-context, OOC）虚假信息的检测问题，此类信息通过将真实图像与误导性文本描述配对构建虚假叙事，其核心挑战在于多模态对齐而非图像伪造鉴定。由于尼泊尔尚无公开的OOC基准数据集，研究者提出了首个面向尼泊尔语主导的多语言OOC基准NepOOC，包含1,090个图像-文本对（545个原始样本，545个OOC样本），并依据五类典型错误（虚构、误标、时间错配、地理错配、身份错配）进行标注，标注者间一致性Kappa值达0.84。系统评估表明，在当前数据规模下，仅依赖文本语义即可实现优异性能：纯文本mBERT模型在宏平均F1得分上达到94.65±0.20%，与最佳多模态模型（ResNet-50+mBERT）无统计学差异（McNemar检验中位p值=1.000，5个种子中无一显著，α=0.05）。相比之下，仅使用图像的模型表现接近随机水平（33–50%），提示未来进展更应依赖数据集扩展而非模型架构复杂化或区域特化设计。因此，解决方案的关键在于利用高质量的多语言文本语义表征，而非依赖复杂的多模态融合机制。
链接: https://arxiv.org/abs/2608.19212
作者: Sanjeev Khatiwada
机构: Independent Researcher (独立研究员)
类目: Computation and Language (cs.CL); Computer Vision and Pattern Recognition (cs.CV)
备注: 12 pages, 5 figures
Abstract:Out-of-context (OOC) misinformation pairs authentic images with misleading captions to construct false narratives without image manipulation, making detection a problem of multimodal alignment rather than image forensics. Despite the prevalence and consequences of OOC misinformation in Nepal, no public benchmark exists for Nepali. We introduce NepOOC, the first publicly available Nepali-dominant multilingual OOC benchmark, comprising 1,090 image-caption pairs (545 pristine, 545 OOC) annotated across five typologies (fabricated, miscaptioned, temporal mismatch, geographic mismatch, identity mismatch) with inter-annotator agreement kappa = 0.84. Systematic evaluation of five multimodal architectures alongside text-only and image-only baselines reveals that caption semantics appear sufficient for strong performance at the current dataset scale. A text-only mBERT model achieves 94.65+/-0.20% Macro-F1, statistically equivalent to the best multimodal system (ResNet-50+mBERT, 94.65+/-0.20%; McNemar median p = 1.000, 0/5 seeds significant at alpha = 0.05). Image-only models perform near chance (33-50%), while training-size scaling suggests that dataset expansion is a more direct path to progress than architectural sophistication or regional specialisation.
[NLP-59] Represented but Ignored: A Causal Account of Prosodic Underuse in Audio-Language Models
【速读】： 该论文旨在解决大音频语言模型（audio-LLM）在表达性语音理解中的关键瓶颈问题，即模型虽能感知并内部表征语调信息（prosody），却无法有效将其体现在最终输出中。其核心挑战在于：现有行为评估难以区分模型失败是源于声学信息丢失、内部解释错误，还是对已有表征的使用失效。为此，作者提出一种分阶段探针梯度（stage-specific probe ladder）方法，用于定位不同阶段的故障模式。研究发现，在四个仅执行理解任务的audio-LLM中，语调信息通常在音频路径中得以保留，并可在大模型后期状态中解码，但仅部分反映于最终响应。通过针对性的隐藏状态干预实验，验证了该潜在表征的因果作用——单次在特定层的编辑即可显著引导模型决策向被抑制的语调判断方向偏移，表明该信号具有可恢复性且存在于一个低维子空间内。特征层面分析进一步揭示，高重要性特征与已知携带语调信息的声学线索高度一致。因此，研究识别出系统性瓶颈并非语调感知本身，而在于将正确感知到的语调信息有效“使用”于输出生成的过程。
链接: https://arxiv.org/abs/2608.19211
作者: Linkai Peng,Baorian Nuchged
机构: University of Connecticut (康涅狄格大学); The University of Texas at Austin (德克萨斯大学奥斯汀分校)
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI); Sound (cs.SD); Audio and Speech Processing (eess.AS)
备注:
Abstract:Human speech is richly expressive, with prosody carrying linguistic and emotional information beyond the lexical content. A capable large audio-language model (audio-LLM) should therefore support expressive speech understanding, not only transcribing what was said but also interpreting how it was said. Yet behavioral evaluations alone cannot reveal why a model fails on prosodic input. An error may reflect loss of acoustic information, incorrect internal interpretation, or failure to use a representation that is already available inside the model. We introduce a stage-specific probe ladder for localizing these failure modes in audio-LLMs. Across four understanding-only audio-LLMs, prosodic information is usually preserved in the audio path and decodable in late LLM states. Yet it is only partially expressed in the model’s final response. We test the causal status of this latent representation with targeted hidden-state interventions. Every intervention shifts the answer distribution in the predicted direction, and in most model–task cells a single edit at the relevant layer is sufficient to drive the model toward the suppressed prosodic decision, though this recovery is directional rather than a selective restoration of the correct class. Feature-level analysis further suggests that this recoverable signal can be expressed through a small subspace. Some of the highest-attribution features in this analysis align with acoustic cues known to carry prosodic information. Within the matched-content contrasts we test, these results locate the recurring bottleneck not in perceiving prosody but in using it. Models that hear and correctly represent a prosodic cue can still fail to express it in their answers.
[NLP-60] When Irrelevant Text Matters: Affine Margin Shifts in Multimodal Large Language Models
【速读】： 该论文旨在解决多模态大语言模型（Multimodal Large Language Models, MLLMs）在视觉语义任务中受到无关文本上下文干扰的问题，尤其关注此类辅助性文本对模型决策的系统性偏移影响。其核心解决方案在于构建一个二元视觉判断框架，通过控制提示结构并动态调整无关上下文输入，系统性地考察上下文对模型输出的影响。关键创新点在于提出以“决策边界”（decision margin）——即二元候选答案间对数概率差——作为量化分析工具，揭示出无关上下文引起的偏差并非随机噪声，而是呈现出可建模的仿射变换规律：条件化决策边界与无上下文情形下的边界之间存在稳定的线性关系。这一几何规律表明，无关上下文实质上以可估计的方式扭曲了模型的偏好分布。进一步地，通过拟合该仿射变换的参数，可分别度量模型对视觉信息的保留能力（visual commitment preservation）和回答方向性偏差（directional answer bias），从而在决策边界层面提供了对无关上下文效应的诊断机制，为未来提升模型在噪声上下文环境下的鲁棒性研究奠定了理论基础。
链接: https://arxiv.org/abs/2608.19208
作者: Yinfeng Wang,Zhiyuan Yao,Zheren Fu,Lei Zhang,Zhendong Mao
机构: University of Science and Technology of China (中国科学技术大学)
类目: Computation and Language (cs.CL); Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Multimodal large language models (MLLMs) are frequently exposed to auxiliary textual context, the impact of which on visually grounded tasks remains underexplored. In this paper, we investigate the influence of task-irrelevant context by formulating it as a controlled intervention within a binary visual judgment framework. By maintaining an invariant prompt structure while varying auxiliary inputs, we observe that irrelevant text consistently biases model predictions across diverse benchmarks. To move beyond performance metrics, we characterize this sensitivity through a decision margin defined by the log-probability difference between binary candidates. Our analysis reveals a robust geometric regularity: contextconditioned margins follow a consistent affine transformation of their context-free counterparts. This finding demonstrates that irrelevant context does not manifest as unstructured stochastic noise but as a estimable distortion of model preference. We further interpret the fitted affine parameters as metrics for visual commitment preservation and directional answer bias. These findings provide a margin-level diagnostic view of irrelevant-context effects in MLLMs and offer a basis for future studies on noisy-context robustness
[NLP-61] Compliance Capability and Conflict: Benchmarking Multimodal LLM s under System Messages
【速读】： 该论文旨在解决多模态大语言模型（Multimodal Large Language Models, MLLMs）在生产部署中对系统消息（system message）遵从性评估不足的问题。现有基准测试要么仅评估纯文本情境下的约束遵守情况，要么将约束嵌入用户输入，导致在多模态场景下对系统消息遵循行为缺乏系统性度量；同时，也未明确模型在遵守指令时是否牺牲了基础的视觉-语言理解能力。为此，本文提出VSysBench，一个基于MMVet-v2构建的基准测试框架，将约束划分为5大类共22个子类，涵盖从视觉语境中的文本指令到完全基于视觉的约束，并为每项约束设计一个语义不一致的对照项以检验指令优先级的敏感性。该基准通过联合满意度率（Joint Satisfaction Rate, JSR）和跨约束敏感性（Cross-Constraint Sensitivity, CCS）两个维度，同步评估响应在指令遵从性与答案正确性上的表现。实验结果表明：系统消息的引入显著降低了模型的基础任务准确率；对于开源模型，当用户指令与系统消息冲突时，遵从性迅速崩溃，而顶级闭源模型则保持稳定；此外，所有模型在完全视觉依赖的约束类别上表现最差，表明此类约束是当前MLLMs面临的最大挑战。
链接: https://arxiv.org/abs/2608.19207
作者: Juan Yeo,Geewook Kim
机构: NAVER Cloud AI(NAVER云AI); KAIST AI(韩国科学技术院人工智能); Seoul National University(首尔国立大学)
类目: Computation and Language (cs.CL)
备注:
Abstract:Production deployments of Multimodal Large Language Models (MLLMs) increasingly rely on system messages to govern model behavior. Yet existing benchmarks either evaluate constraints in text only or embed them into the user turn, leaving system-message adherence in multimodal contexts largely unmeasured; they also leave open whether compliance comes at the cost of foundational vision-language capabilities. We introduce VSysBench, a benchmark built on MMVet-v2 that organizes constraints into 5 main categories and 22 sub-categories, ranging from textual directives in visual contexts to fully vision-grounded ones, each paired with a misaligned counterpart that stress-tests the instructional hierarchy. VSysBench scores each response jointly along two axes, constraint compliance and answer correctness, via the Joint Satisfaction Rate (JSR) and Cross-Constraint Sensitivity (CCS). Across 16 MLLMs, we find that imposing system messages substantially erodes base task accuracy, that compliance collapses under user conflict for open-weight models while remaining stable for top proprietary ones, and that vision-grounded constraints are the hardest category for every model.
[NLP-62] Asymmetric Attention Heads: Structured Head-Wise Context Allocation for Transformer Attention
【速读】： 该论文旨在解决标准多头注意力（Multi-Head Attention, MHA）机制中所有注意力头均使用相同全局因果上下文窗口的局限性，尽管不同注意力头在实际任务中可能承担不同的上下文角色——部分头依赖局部词汇或句法信息，而另一些头则需建模长距离关系（如实体交互、话语连结或状态变化）。为此，论文提出非对称注意力头（Asymmetric Attention Heads, AAH），其核心在于将上下文长度作为每个头或头组的显式分配变量，通过基于特征统计量对注意力头进行分组，并构建层次化分组结构，为各头分配局部因果窗口，同时保持与标准MHA一致的扁平输出接口。实验表明，在4096个标记的种子0设置下，多种AAH风格的局部分配变体均取得比纯全注意力更低的验证损失；短预算消融实验进一步揭示，稳定的局部分配策略与头-窗口分配结构至关重要，而固定或局部控制基线在某些情况下可与自适应层级结构相媲美。作者将AAH解释为一种结构化的头级上下文分配机制，引入**注意力覆盖比率（Attention Coverage Ratio, ACR）**作为选择性窗口路由的诊断指标，以支持模型可解释性分析。
链接: https://arxiv.org/abs/2608.19203
作者: Zimu Zhao
机构: 未知
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
备注: 30 pages, 4 figures
Abstract:Standard multi-head attention (MHA) gives every head the same full causal context span, although heads can serve different contextual roles. Some heads may rely mainly on nearby lexical or syntactic context, while others may depend on longer-range relations such as entity interactions, discourse links, or state changes. We present Asymmetric Attention Heads (AAH), a head-wise context- allocation framework that treats context length as an explicit per-head or per-group allocation variable. AAH groups heads using feature-derived statistics, organizes these groups hierarchically, and assigns causal local windows while preserving the standard flat MHA output interface. In 4096- token seed-0 experiments, several AAH-style local-allocation variants achieve lower validation loss than pure full attention. Short-budget ablations show that stable local allocation and head-window assignment structure matter, while fixed/local controls can be competitive with adaptive hierarchy. We interpret AAH as a structured head-wise context-allocation mechanism for quality and analysis, with Attention Coverage Ratio (ACR) reported as a selected-window routing diagnostic
[NLP-63] Active Inference as Context Acquisition for AI Agents
【速读】： 该论文旨在解决交互式人工智能（AI）代理在获取正确上下文时面临的效率难题：当用户遗漏约束条件、偏好、文件或任务变量时，代理需在采用默认假设与消耗资源进行澄清性提问、检索、工具调用或提示试验之间做出权衡。为此，论文将这一决策过程形式化为主动推理（active inference）框架下的上下文获取问题。其核心解决方案在于构建一个双层优化结构：内层推理步骤基于观测更新对潜在任务状态的信念分布，外层决策则选择能够最小化预期自由能（expected free energy）的下一步动作——包括上下文获取动作、任务执行动作或终止动作，同时考虑计算成本。在确定性场景下，认知项退化为期望信息增益，可按令牌成本进行归一化。作者通过构建最优提问机制（Optimal Question Asking, OQA） 实现该框架，采用精确后验分布与动态规划最优解作为基准，在包含25至300个候选项的二元及多分类任务上对前沿语言模型进行了评估，并进一步研究了生成前澄清与受限令牌预算下的自动提示优化策略。该方法具有模型无关性，将主动推理视为AI代理上下文获取层的设计范式。
链接: https://arxiv.org/abs/2608.19202
作者: Sanchayan Dutta,Sai Niranjan Ramachandran,Suvrit Sra
机构: University of California, Davis; Technical University of Munich (慕尼黑工业大学)
类目: Artificial Intelligence (cs.AI); Computation and Language (cs.CL); Machine Learning (cs.LG)
备注:
Abstract:Interactive AI agents must acquire the right context as efficiently as possible. When a user omits a constraint, preference, file, or task variable, an agent can proceed with a default assumption or spend tokens on a clarifying question, retrieval call, tool call, or prompt trial. We formulate this tradeoff as active inference for context acquisition. An inner inference step updates beliefs over a latent task state, and an outer decision selects the next context action, task action, or stop action to minimize expected free energy under cost. In deterministic settings, the epistemic term reduces to expected information gain, optionally normalized by token cost. We instantiate the framework in Optimal Question Asking (OQA), with exact posteriors and a dynamic programming oracle, and benchmark frontier language models on binary and multiway categorical tasks from 25 to 300 candidates. We also study clarification before generation and automated prompt optimization under token budgets. The formulation is model-agnostic and views active inference as a design principle for the context-acquisition layer of AI agents.
[NLP-64] ransformer Models for Text Summarization: A Comparative Study of BART BERT and RoBERTa
【速读】： 该论文旨在解决自动文本摘要（Automatic Text Summarization, ATS）中如何有效生成高质量、信息密集且语言流畅的摘要这一核心问题，尤其聚焦于在复杂文档理解与生成能力上的挑战。其解决方案的关键在于利用基于Transformer架构的大规模预训练模型（如BERT、RoBERTa和BART），通过深度上下文建模能力实现对输入文本的精准语义理解，并结合先进的预训练策略（如掩码语言建模、自回归生成等），显著提升提取式（extractive）与抽象式（abstractive）摘要任务的表现，从而在保持关键信息完整性的同时增强生成内容的连贯性与自然性。
链接: https://arxiv.org/abs/2608.19200
作者: Daisy Aptovska,Vinayak Elangovan
机构: 未知
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
备注: 1- pages
Abstract:Text summarization refers to the task of condensing a document into a shorter version while preserving its key information. Automatic text summarization (ATS), driven by advancements in natural language processing (NLP), has developed rapidly in recent years. ATS methods are commonly categorized by input type (such as single-document or multi-document summarization) and by output type (extractive, abstractive, and hybrid). This article presents a focused review of modern summarization techniques with an emphasis on transformer based models and large language models (LLMs), specifically BERT, RoBERTa and BART. It examines their architectures, pretraining strategies, and their suitability for extractive and abstractive summarization tasks.
信息检索
[IR-0] Projecting BrowseComp-Plus onto ClimbMix: Toward More Realistic Corpora for Agent ic Search
链接: https://arxiv.org/abs/2608.20317
作者: Sahel Sharifymoghaddam,Lingwei Gu,Yijun Ge,Jimmy Lin
类目: Information Retrieval (cs.IR)
备注:
Abstract:The BrowseComp-Plus benchmark disentangled the evaluation of agentic search by replacing opaque web search with a fixed corpus, so that an agent’s role can be separated from the retriever’s. That corpus, however, holds only about 100K documents and was assembled from the supporting documents of the benchmark’s own queries plus mined hard negatives, so the evidence and the distractors were both selected per query. We introduce \textBrowseComp-Plus_\textCM , which keeps the BrowseComp-Plus questions but relocates their evidence to ClimbMix, a 400B-token, 553M-document mixture of web text released by NVIDIA for pre-training language models and built without reference to any benchmark. Our main contribution is the projection pipeline that makes this possible: it decomposes each question into atomic reasoning hops and grounds every hop in the new corpus, retaining a question only when automatic verification, an independent agent, and human review all confirm that every hop is supported. The pipeline is dataset-agnostic and applies to any benchmark whose questions decompose into verifiable facts. Applied to the 830 BrowseComp-Plus test questions, our pipeline yields 57 fully grounded questions with question-level relevance judgments. Projection shifts the difficulty onto retrieval, as the strongest agent we evaluate loses five points of answer accuracy but sees its evidence recall fall from 84.3% to 21.4% while issuing 63% more search calls. As the first of a series of projections, we release the pipeline, the benchmark, and our analyses at this https URL.
[IR-1] What Makes a Good Fiqh Retriever? Answer Retrieval for Arabic Islamic Jurisprudence
链接: https://arxiv.org/abs/2608.20246
作者: Somaya Eltanbouly,Heba Sbahi,Samer Rashwani,Abdessalam Bouchekif,Mutaz al-Khatib,Shahd Gaben,Mohammed Ghaly
类目: Information Retrieval (cs.IR)
备注:
Abstract:Retrieval-Augmented Generation is used for Islamic question answering, but most systems are evaluated end-to-end, making retrieval failures difficult to isolate from generation failures. We study answer-bearing retrieval for Arabic fiqh, where a passage is relevant only if it states the ruling required by the question. We build a retrieval test collection for Arabic fiqh and use it to evaluate dense, lexical, hybrid, fine-tuned, and madhhab-aware retrieval strategies. The best retriever achieves 0.524 MRR@5, while fine-tuning improves performance to 0.553. Hybrid retrieval provides limited gains for strong models, whereas madhhab-aware filtering more than doubles MRR@5 on school-specific questions. We further present an error analysis showing that the main challenge is distinguishing answer-bearing passages from topically similar passages that do not contain the requested ruling.
[IR-2] Daedalus-150M: A Convolution-Attention Hybrid Designed for CPU Inference
链接: https://arxiv.org/abs/2608.20210
作者: Christos Koutsiaris
类目: Information Retrieval (cs.IR); Artificial Intelligence (cs.AI); Computation and Language (cs.CL); Machine Learning (cs.LG)
备注: 8 pages, 2 figures, 10 tables. Code, weights and the measurement artefacts behind every number: this https URL
Abstract:Small language models are usually built like large ones and then squeezed onto a CPU afterwards. We did the opposite: we fixed the target first, one user, one token at a time, 4-bit weights, ordinary CPU, and chose the architecture to suit it. The result keeps full attention in only 6 of its 18 blocks. The other 12 use short convolutions whose memory is two timesteps wide no matter how long the conversation gets, so two thirds of the network never re-reads a growing cache. Trained from scratch on 59.9B tokens, the model scores 47.31 on a five-task benchmark against a bar of 42.20 that was fixed before training began. It beats GPT-2 124M, Pythia-160M, OPT-125M and GPT-neo-125M, all trained on three to six times more data, and exceeds MobileLLM-125M’s published score despite that model seeing a trillion tokens. Validation bits-per-byte is 0.8685. To check the architecture rather than the training recipe, we trained a conventional all-attention model of the same size on the same data, and wrote down the winning condition before scoring either. The hybrid won the chosen quality metric by 0.81%, matched it on downstream tasks, produced a 6.3% smaller 4-bit file, and decoded 1.76x faster at 2048 tokens of context, 2.08x against an external model of similar size. In every measurement the speed advantage is near zero at an empty context and grows with length, which is what the mechanism predicts and what a merely leaner model would not show. A simple bandwidth calculation predicts only 1.17x, so memory volume alone does not explain the gap. We also report what did not work: an unmitigated 4-bit quality cost, roughly half the convolution channels ending up inert and impossible to remove, and a vocabulary larger than this model size warrants. Comments: 8 pages, 2 figures, 10 tables. Code, weights and the measurement artefacts behind every number: this https URL Subjects: Information Retrieval (cs.IR); Artificial Intelligence (cs.AI); Computation and Language (cs.CL); Machine Learning (cs.LG) ACMclasses: I.2.7; I.2.6 Cite as: arXiv:2608.20210 [cs.IR] (or arXiv:2608.20210v1 [cs.IR] for this version) https://doi.org/10.48550/arXiv.2608.20210 Focus to learn more arXiv-issued DOI via DataCite (pending registration) Submission history From: Christos Koutsiaris [view email] [v1] Thu, 20 Aug 2026 16:09:43 UTC (18 KB)
[IR-3] SCoRD: Semantic-Assisted Continual Retriever-Reranker Distillation for LLM -Based Recommendation
链接: https://arxiv.org/abs/2608.19998
作者: Seunghyun Baek,Gyuseok Lee,Seunghan Lee,Wonbin Kweon,Dong Wang,SeongKu Kang
类目: Information Retrieval (cs.IR)
备注:
Abstract:Recommendation systems increasingly adopt a two-stage pipeline, where an ID-based retriever retrieves candidates and an LLM-based reranker refines their rankings. To improve retrieval quality, reranker-to-retriever distillation is commonly used to transfer the reranker’s knowledge to the retriever. For practical deployment, however, this pipeline must continually adapt to evolving interests and incoming interactions. A naive solution is to repeatedly update the LLM reranker and distill its latest knowledge, but this incurs prohibitive costs. Updating the retriever alone is cheaper, but its limited capacity makes adaptation from sparse data difficult. We propose SCoRD, a continual knowledge distillation framework for LLM-based reranking pipelines under a non-stationary data stream. SCoRD introduces a semantic reasoning assistant that distills the LLM’s ability to infer underlying user intents into reusable intent-level guidance. It selectively distills reranker knowledge to the retriever on low-confidence sequences, guides retriever-only updates without repeated LLM inference, and feeds retriever-derived representations and intent-drift signals back to the reranker. Experiments on real-world datasets show that SCoRD enables effective and efficient retriever-reranker co-adaptation.
[IR-4] Do Sequential Recommendation Benchmarks Really Require Higher-Order Sequence Modelling? RECSYS2026
链接: https://arxiv.org/abs/2608.19833
作者: Aleksandr V. Petrov,Praveen Chandar,Paul N. Bennett,Hugues Bouchard,Mounia Lalmas
类目: Information Retrieval (cs.IR)
备注: Accepted at the 20th ACM Conference on Recommender Systems (RecSys 2026)
Abstract:Sequential recommenders increasingly use language-model architectures designed to capture complex, context-dependent interactions. Yet it remains unclear whether widely used benchmarks actually require this modelling capacity. We investigate this question using two simple, recency-weighted pairwise probes that do not learn higher-order sequence representations: Sequential Rules (SeqRules) and our Probabilistic Collaborative Transition Model (PCTM). Using the evaluation protocol of eSASRec, at least one probe exceeds our eSASRec reproduction by 15-38% on three Amazon datasets and by 4.4% on MovieLens-1M, but trails it by 27.3% on MovieLens-20M. On the four remaining datasets, at least one probe also outperforms our sampled-softmax SASRec reproduction by 9-28%, suggesting that these widely used benchmarks are poorly suited to measuring gains from higher-order sequence modelling. More broadly, comparing Transformer-based models against strong recency-weighted pairwise probes provides a concrete test of whether a benchmark can meaningfully measure gains from higher-order sequence modelling.
[IR-5] raining-Free LLM -Based Recommendation with Post-LLM Item Refinement Using Collaborative Signals CIKM2026
链接: https://arxiv.org/abs/2608.19665
作者: Kyungho Kim,Sunwoo Kim,Geon Lee,Shinhwan Kang,Sojeong Kim,Liam Collins,Bhuvesh Kumar,Donald Loveland,Kijung Shin
类目: Information Retrieval (cs.IR)
备注: Published as a conference paper at CIKM 2026 (short)
Abstract:Large language models (LLMs) have shown promise for training-free recommendation, but LLM-generated user interests are often too broad for fine-grained item retrieval. Existing methods incorporate collaborative filtering (CF) signals in a pre-LLM manner through candidate reranking or prompt augmentation, yielding limited gains. We propose CoRRe, a training-free recommendation framework with a post-LLM paradigm that injects CF signals into LLM-generated item representations, which are later matched with LLM-generated user interests for ranking. Specifically, CoRRe refines the directions of item embeddings using an item-item co-purchase graph and their magnitudes using item popularity. Experiments on real-world datasets show that CoRRe consistently outperforms existing training-free methods and achieves competitive or superior performance compared with training-based methods, without requiring any model training or task-specific fine-tuning.
[IR-6] SSR-GRPO: Integrating Supervision and Semantic IDs into Reinforcement Learning for Dense Retrieval in E-commerce
链接: https://arxiv.org/abs/2608.19595
作者: Guangxin Song,Xing Fang,Mingmin Jin,Jing Wang,Bokang Wang,Zhentao Song,Junjie Bai,Jianbo Zhu
类目: Information Retrieval (cs.IR)
备注:
Abstract:Embedding-based retrieval (EBR) is pivotal in e-commerce search but often struggles with complex semantics. While recent methods often fine-tune large language models (LLMs) for representation learning, they typically lack robust mechanisms for handling complex and implicit semantics. While Retrieval-GRPO (R-GRPO) recently introduced reinforcement learning to dense retrieval, it suffers from noisy top-K candidates due to limited batch sampling and biased relevance assessments caused by using similarly trained LLMs as reward models. To tackle these issues, we propose Supervised Retrieval-GRPO with Semantic Identifiers (SSR-GRPO). Specifically, our method first proposes a dual-perspective framework for relevance assessment. It leverages both Semantic Identifiers (SIDs) produced by quantization learning and dense representation vectors to generate more unbiased relevance scores. Furthermore, leveraging the hierarchical similarity relationships of the generated SIDs, we mine a set of hard negative samples that serve two purposes: (1) to design a masking function integrated into R-GRPO, effectively filtering intra-group noisy samples; and (2) to construct a Retrieval-DPO task composed of positive and negative sample pairs, enabling the model to capture fine-grained semantic distinctions from a pair-wise perspective. By integrating these optimization strategies, we propose SSR-GRPO. Extensive offline and online experiments validate SSR-GRPO’s effectiveness, and it has been deployed on a large-scale e-commerce platform.
[IR-7] wo-sided receptivity to conversational AI agents in online dating: Bilingual survey data from Fledge.Love
链接: https://arxiv.org/abs/2608.19545
作者: Daria Leshchikova,Valentina V. Kuskova,Dmitry Zaytsev,Valerii Klimov
类目: Computers and Society (cs.CY); Information Retrieval (cs.IR)
备注:
Abstract:Autonomous conversational agents and generative-AI features are being added to online dating platforms faster than public evidence about user attitudes can accumulate, and the scarcest evidence concerns the receiving side: how people react when the profiles, messages, or conversation partners they encounter are machine-generated. We release two anonymized survey datasets collected from active users of this http URL, a dating platform serving an international user base. The first (N = 2,617; Russian and English forms) measures receptivity to autonomous conversational agents with a seven-item battery that separates the principal role (deploying one’s own agent) from the counterpart role (encountering someone else’s), plus six ordinal covariates and two auxiliary items. The second (N = 2,894) measures interest in three passive generative-AI features. The release includes model-derived scores for 2,499 complete cases, a bilingual codebook, a documented anonymization pipeline with a k-anonymity audit, executable analysis notebooks, and canonical outputs, supporting reuse in human-AI communication, recommender-systems, and cross-cultural technology-acceptance research.
[IR-8] From Retrieved Context to Runtime Control: Adaptive Compression for Edge-based RAG
链接: https://arxiv.org/abs/2608.19535
作者: Zlatan Feric,Amir Taherin,Yanzhi Wang,David Kaeli
类目: Artificial Intelligence (cs.AI); Computation and Language (cs.CL); Distributed, Parallel, and Cluster Computing (cs.DC); Information Retrieval (cs.IR); Performance (cs.PF)
备注: Accepted to appear in the Proceedings of the ACM AI Leadership Summit 2026. Zlatan Feric and Amir Taherin contributed equally
Abstract:Retrieval-augmented generation (RAG) improves language-model responses by grounding generation in external passages, which comes with overhead: retrieved context lengthens the prompt, increasing prefill work, KV-cache footprint, memory traffic, latency, and energy. Context compression offers a natural remedy by pruning retrieved text before generation. However, state-of-the-art context-compression methods are typically used with a fixed compression budget, or with the rate selected offline and then applied at inference time. This static view ignores both workload variation and the live state of the edge device. On an edge SoC, compression is not free: the compressor itself runs on the same SoC and consumes latency and energy that can offset any generation savings. This paper proposes a vision for telemetry-informed adaptive compression in edge RAG, grounded in experimental evidence. We characterize the compression tradeoff on the NVIDIA Jetson AGX Thor using Llama and Qwen generators, Natural Questions and HotpotQA datasets, and LLMLingua-2 compression. Our measurements show that generation dominates the RAG budget for larger models, reaching roughly 90% of per-query latency and 91% of GPU energy for 7B-8B generators. Exploring the impact of the compression rate reveals an adaptive operating region: mild compression can miss energy opportunities, and overly aggressive compression can hurt inference quality. Intermediate compression can reduce GPU energy by up to 53.2%, and SoC energy by up to 48.2%, with negligible quality loss. We argue for runtime policies that dynamically manage compression, guided by workload features and edge telemetry. Comments: Accepted to appear in the Proceedings of the ACM AI Leadership Summit 2026. Zlatan Feric and Amir Taherin contributed equally Subjects: Artificial Intelligence (cs.AI); Computation and Language (cs.CL); Distributed, Parallel, and Cluster Computing (cs.DC); Information Retrieval (cs.IR); Performance (cs.PF) Cite as: arXiv:2608.19535 [cs.AI] (or arXiv:2608.19535v1 [cs.AI] for this version) https://doi.org/10.48550/arXiv.2608.19535 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
[IR-9] Automated Summarization of Financial News Using Large Language Models and Retrieval-Augmented Generation: An Early Empirical Study (Fall 2023)
链接: https://arxiv.org/abs/2608.19526
作者: Pranav Chandaliya
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI); Information Retrieval (cs.IR)
备注: 17 pages, 1 figure, 6 tables. Research conducted Fall 2023 at George Washington University; manuscript prepared for public release in 2026
Abstract:Stock market analysts and investors face a daily challenge: too much financial news, too little time. Manually reading and synthesizing hundreds of company-specific articles is impractical, yet missing key information can directly affect investment decisions. This project, conducted at George Washington University in Fall 2023, explores whether Large Language Models can automate this process reliably. We built a pipeline that pulls news articles from the News API, company background from Wikipedia, and stock price data from Yahoo Finance for ten major companies (AAPL, MSFT, GOOGL, AMZN, META, TSLA, JPM, NVDA, WMT, DIS). Because LLMs cannot directly process numerical tables, we developed a simple but effective template that converts stock data into natural language narratives. We then tested two summarization approaches (Summarize Chains and Retrieval-Augmented Generation with FAISS) across three open-source models (Falcon-7B-Instruct, DistilBART-CNN-12-6, BART-Large-XSum) for news, and GPT (text-davinci-003) for stock summaries. Falcon-7B with Summarize Chains gave the best results, covering all news events accurately and coherently. RAG, while promising in theory, caused severe repetition in Falcon and hallucinated facts in BART-Large when k was large. Both LLM-based approaches outperformed a simple Lead-3 baseline on ROUGE-1. We also built a Streamlit dashboard for interactive stock visualization. The work was done in Fall 2023, before RAG-based financial tools became widespread, and the failure modes we document, particularly hallucination under RAG in smaller models, remain relevant today.
[IR-10] HARP: Hierarchical Adaptive Ranking with Preference-Adaptive Fusion for Query-Based CVE Prioritization
链接: https://arxiv.org/abs/2608.19430
作者: Haochen Liu,Zhengzhang Chen,Haoyu Wang,Yanchi Liu,Jundong Li,Haifeng Chen
类目: Information Retrieval (cs.IR); Computation and Language (cs.CL); Cryptography and Security (cs.CR)
备注:
Abstract:Vulnerability prioritization is inherently preference dependent, since the same CVE can receive different remediation priority under different operational preference scenarios. Existing scoring systems and ranking methods typically assume a fixed criterion. In practice, organizations already operate under a preference scenario, but this preference is often implicit and difficult to express as a written prompt instruction, while triage queries usually do not encode it. Past validated triage cases under the current scenario are more readily available. We study query-based CVE prioritization in this setting and propose HARP, a graph-grounded multi-view framework that ranks candidates from a natural-language query together with a support bank of historical labeled examples from the current preference scenario, without requiring an explicit textual summary of that scenario. HARP retrieves evidence from a vulnerability knowledge graph, scores candidates with policy-conditioned global, enterprise, and user views, and fits view-fusion weights from sampled supports. Experiments across three preference scenarios and multiple backbone LLMs show that HARP outperforms multiple baselines, expressing our method’s effectiveness.
[IR-11] Quantization Beyond Uniform Bit Allocation VLDB2026
链接: https://arxiv.org/abs/2608.19388
作者: K. S. Sreeramji,Sabyasachi Basu,Ravishankar Krishnaswamy,Kirankumar Shiragur,Yujia Wang
类目: Information Retrieval (cs.IR); Databases (cs.DB)
备注: Accepted at the 2nd Workshop on Vector Databases (VecDB), part of 52nd International Conference on Very Large Data Bases (VLDB 2026)
Abstract:Quantization is a fundamental technique to handle the growing sizes of embeddings generated by modern models. Existing quantization schemes are largely embedding agnostic and allocate bits uniformly across dimensions. However, recent models produce embeddings with significant geometric structure. In this work, we investigate whether a variable bit allocation scheme can improve quantization quality under a fixed memory budget. We propose a simple variable bit allocation framework that partitions an embedding into contiguous buckets and allocates storage non-uniformly across them. Using a greedy allocation strategy, we instantiate this framework for both Product Quantization (PQ) and Scalar Quantization (SQ). We perform a series of experiments on embeddings known to have the Matryoshka property (MRL), and consistently observe that non-uniform allocations outperform uniform baselines at identical storage budgets. The largest improvements occur in the low-bit regime, where uniform allocation is particularly inefficient for MRL embeddings. At the same compression rates, variable allocation improves recall by up to 8% for PQ and up to 18% for SQ. Our results suggest a new direction for structure-aware compression and indexing techniques for large-scale retrieval systems.
[IR-12] CrossQ: Task-Aligned Cross-Token Conditional Quantization for Late Interaction Retrieval
链接: https://arxiv.org/abs/2608.19204
作者: Rohit Kumar Salla,Manoj Saravanan,Ramya Manasa Amancherla
类目: Information Retrieval (cs.IR)
备注:
Abstract:Late-interaction retrievers like ColBERT achieve high quality but suffer from large multi-vector indices. Standard compression minimizes token reconstruction error, while ranking depends critically on preserving scores of sparse “winner” tokens. We introduce CrossQ, which adaptively improves effective token fidelity within documents by conditioning token codes on lightweight document context computed at indexing time (but not stored). CrossQ is trained with ranking-aligned objectives that preserve candidate score distributions and protect hard-negative margins. At 2 B/token, CrossQ improves MRR@10 by +0.010 over the strongest strictly footprint-matched quantization baseline and by +0.012 over the strongest candidate-matched system reference. On a nine-dataset BEIR subset, CrossQ improves average nDCG@10 by +0.009 at 4 B/token over the strongest candidate-matched system reference. At 4 B/token, CrossQ achieves 64x raw token-storage reduction, approximately 61x including metadata and approximately 58x under conservative padding/alignment accounting. At 8 B/token, CrossQ with light fine-tuning retains approximately 98% of full-precision ColBERT MRR@10, improving the footprint-quality tradeoff for memory-constrained late-interaction retrieval.
[IR-13] Automatic bioinformatic software named entity recognition from literature
链接: https://arxiv.org/abs/2608.19201
作者: Hao Xuan,Rithvij Pasupuleti,Ben Liu,Haishuo Sun,Jun Zhang,Zijun Yao,Cuncong Zhong
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI); Information Retrieval (cs.IR); Quantitative Methods (q-bio.QM)
备注:
Abstract:Bioinformatics software and databases are essential components of modern life science research, yet their mentions in the scientific literature are often inconsistent and difficult to systematically identify at scale. The lack of a comprehensive and up-to-date catalog of bioinformatics resources hinders efforts toward automated biomedical knowledge extraction and streamlined data analysis. Here we present SNAIL, a hybrid named entity recognition framework designed to automatically identify bioinformatics software and database (SW/DB) names from biomedical texts. SNAIL integrates complementary lexical and semantic modeling strategies. The lexical component captures orthographic patterns and contextual cues characteristic of SW/DB names, while the semantic component leverages contextual embeddings generated by transformer-based language models such as SciBERT, combined with an explicit token-masking strategy to enhance entity-focused representations. A large training corpus was constructed automatically through a hybrid pipeline that integrates citation-hinted extraction with large language model-assisted distillation. Evaluation on two independent benchmark datasets and real-world research articles demonstrates that SNAIL substantially outperforms existing approaches, including domain-specific methods such as bioNerDS2 and general-purpose large language models such as ChatGPT, Gemini, Grok and Claude. Applying SNAIL to large-scale literature analysis further reveals distinct journal-level preferences across bioinformatics subfields. These results demonstrate that SNAIL provides an accurate and scalable solution for identifying bioinformatics resources in scientific texts and enables systematic meta-analysis of tool usage and research trends.
[IR-14] A Virtual Member of a Community of Practice for the Society of Petroleum Engineers: From Prototype to Deployment
链接: https://arxiv.org/abs/2608.19199
作者: John Boden,Joshua Eckroth,Dayne Freitag,Skyler Gipson,Johnathan Keefe,Karen Myers,Eric Schoen,Pedro Sequeira,Reid Smith,Michael Wessel
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI); Information Retrieval (cs.IR)
备注: 7 pages
Abstract:We describe the evolution of a virtual assistant, called ATHENA, designed to support the capture, retrieval, and dissemination of knowledge for members of a Community of Practice (CoP) related to the Oil and Gas sector. An evaluation of a first prototype involving 75 professionals from the Society of Petroleum Engineering (SPE) showed that ATHENA dramatically improved both their productivity and performance equality on a set of realistic well-planning tasks compare to their use of a state-of-the-art RAG baseline system. However, the evaluation also identified areas for improvement. This paper describes technical advances to our first prototype in the areas of multi-document retrieval, support for answer validation, and more focused proactive dissemination. Evaluation results show that this enhanced version of ATHENA provides better support for completing knowledge-intensive tasks related to well planning than does a state-of-the-art baseline. ATHENA has been integrated into the SPE Research Portal and is being deployed for use by the society’s membership.
人机交互
[HC-0] From Agent Behaviour to Agent -Friendly Documentation: An Empirical Study of How Coding Agents Discover Read and Write Technical Documentation
链接: https://arxiv.org/abs/2608.20195
作者: Zhijun Gao,Jing Chen
类目: oftware Engineering (cs.SE); Artificial Intelligence (cs.AI); Human-Computer Interaction (cs.HC)
备注: 14 pages, 1 figure, 10 tables
Abstract:Technical documentation is written for human developers, but an increasing share of software changes is now authored by autonomous coding agents. Which documents they consult, when, and what follows remain unknown. We conduct a behaviour-grounded study of agent-documentation interaction across two public datasets: 557 agentic coding sessions from SWE-chat, yielding 94,813 development events including 3,033 documentation interactions; and 33,097 agentic pull requests from AIDev, with 690,260 classified file-level change records. Four findings challenge current documentation practice. First, agents’ documentation work is dominated by agent-facing artefacts: instruction files and working notes account for 60.5% of all documentation interactions, versus 10.6% for classical technical documentation and 1.3% for API references. Second, the link between consultation and code editing is unresolved: the adjacent transition probability is 0.002 and the unadjusted three-event lift 1.05, whereas a stage-adjusted model places it above unity (OR 1.33 [1.09, 1.62]); documentation creation is elevated unadjusted (lift 1.67) but its adjusted interval includes unity. Third, no explicit documentation-based validation sequence was observed, and consultation is associated with less immediate testing (lift 0.23, cluster CI 0.08-0.45; adjusted OR 0.39 [0.25, 0.60]). Fourth, consultation is self-initiated (70.2%) far more often than failure-driven (7.5%), and documentation trails code: among multi-commit pull requests changing both, code is touched first 4.7x more often. From these traces we derive a descriptive model of agent-documentation interaction as a two-lobed cycle rather than a linear journey, and show that two widely assumed properties of “agent-friendly” documentation - actionability and verifiability - lack consistent behavioural support. We release our pipeline, coding scheme, and event-level data.
[HC-1] Navigating and Retrieving Information in Immersive Model-Based Design Reviews: An Exploratory Study
链接: https://arxiv.org/abs/2608.20128
作者: Victor Romero,Romain Pinquié,Frédéric Noel
类目: Human-Computer Interaction (cs.HC)
备注:
Abstract:Digital engineering uses many models from different perspectives, creating a connected set of digital artefacts across a product’s life cycle. Designers seeking a holistic view must navigate numerous models and views, requiring domain-specific software, languages, and representations. This can lead to getting lost in scattered information and the cognitive burden of mentally integrating details across diagrams. To overcome these issues, we developed the virtual environment GraphXplore. GraphXplore enhances perceptual and conceptual integration by linking all relevant visual items from different perspectives into an interactive, layered 3D graph displayed in virtual reality, providing a holistic view of the system. We compared GraphXplore with a conventional on-screen setup using a PowerPoint slide deck with model screenshots viewed on a desktop PC. In an experiment with N=33 volunteers (mainly industrial product design postgraduates and professors), we conducted a baseline usability study focused on fundamental information retrieval tasks for model-based design comprehension, as identifying basic model elements is the fundamental prerequisite in design reviews. Our findings indicate that for simple retrieval tasks, the correctness of answers, completion time, recall score, and perceived confidence are comparable in both environments. However, the virtual environment demonstrated practical advantages, achieving a “good” average System Usability Scale (SUS) score of 73.1 compared to the slide setup’s borderline score of 66.4. Furthermore, GraphXplore users reported a lower perceived cognitive workload (mean NASA-TLX score of 43.8) compared to the traditional setup (50.4). Future work will enhance this experiment to yield empirical results across massive industrial datasets, refine GraphXplore, and extend to new design review objectives.
[HC-2] What Do Visualization Instructors Want Students to Learn? Introducing a Concept Inventory for Visualization Design
链接: https://arxiv.org/abs/2608.20090
作者: Medina Lamkin,Heer Patel,Sayamindu Dasgupta,Leilani Battle
类目: Human-Computer Interaction (cs.HC)
备注: 11 pages, 5 figures, submitted to VIS 2026
Abstract:The term “visualization design” encompasses multiple concepts and skills that go well beyond current assessments of graphical perception and visualization literacy. In the context of education, what exactly should a student be able to do if they “know” visualization design? To answer this question, we draw on existing methodology from the field of education to propose a concept inventory for visualization design, i.e., a theoretical model capturing the most important concepts and skills commonly associated with visualization design. We initially draft the concept inventory using a qualitative analysis of course objectives from visualization course syllabi. Then, we iteratively refine the concept inventory by soliciting feedback from instructors through semi-structured interviews. Based on our experiences in developing the concept inventory, we reflect on open questions and future research directions in visualization education, such as developing assessments for visualization design (similar to those for visualization literacy) and providing automated assistance for learning and teaching visualization design. Our supplemental materials are available at this https URL.
[HC-3] Designing Human-mediated AI Guidance: Ready Together for Personalized Family Emergency Preparedness
链接: https://arxiv.org/abs/2608.19950
作者: Nini Kurashvili,Yana Ivanchenko,Greta Schiavo,Cansu Koyuturk,Dimitri Ognibene
类目: Human-Computer Interaction (cs.HC); Artificial Intelligence (cs.AI); Computers and Society (cs.CY)
备注: Accepted for Italian Workshop on Artificial Intelligence for Human-Machine Interaction (AIxHMI 2026), October 6-9, 2026, Perugia, Italy
Abstract:Artificial intelligence (AI) systems are increasingly used across domains to provide personalized information, recommendations, and decision support. However, in some contexts, AI-generated information may not be suitable for direct delivery to the final recipient. Instead, it may need to be interpreted, adapted, and communicated by a human who understands the recipient’s needs, emotional state, and situational context. Human-AI interaction research has given less attention to situations in which a more knowledgeable human acts as an intermediary between an AI system and a less experienced or less informed recipient. We introduce the human-mediated AI guidance framework and explore it through Ready Together, an AI-supported family emergency preparedness system in which parents mediate AI-generated content for their children. The system is designed to provide personalized guidance and support parents in making emergency preparedness more interactive and understandable through guided activities and family-centered learning. The system design was informed by a qualitative, design-oriented research process involving semi-structured interviews and co-design activities. Findings identified challenges in family emergency preparedness, including difficulty discussing emergencies with children, uncertainty about providing appropriate explanations, and a preference for interactive learning activities. These findings informed the design of an interactive prototype, subsequently evaluated through a pilot study and a heuristic evaluation. Participants responded positively to the personalized recommendations and practical activities. Preliminary findings suggest that human-mediated AI guidance may support context-sensitive family preparedness while preserving parents’ responsibility for interpreting, adapting, and communicating AI-generated information.
[HC-4] Evaluating Smart Home Device User Responses to their (Un)Confirmed Privacy Expectations
链接: https://arxiv.org/abs/2608.19873
作者: Tania Khatun,Mahdieh Sheikh Rezaei,Danny Yuxing Huang,Oded Nov,Reza Ghaiumy Anaraky
类目: Human-Computer Interaction (cs.HC)
备注:
Abstract:Users of smart home devices are often unaware of how their devices handle personal data. We examine how revealing these data practices influences user trust, satisfaction, and coping behaviors, including decisions to block device communications. Using Expectation-Confirmation Theory, we conducted two complementary studies to balance ecological validity with experimental control. An in-situ field study used network monitoring to reveal actual device traffic, and an online experiment presented simulated reports with manipulated levels of advertising-related communications. Across both studies, when data practices aligned with user expectations, satisfaction increased, strengthening intentions to continue using the device. Defensive responses, however, followed different pathways: satisfaction predicted willingness to block in the in-situ field study, whereas collection concerns were the primary predictor of blocking in the experiment. Together, these findings show how transparency reshapes attitudes and behaviors among existing smart-home users, underscoring the role of expectation confirmation in real-world, continued-use contexts.
[HC-5] Dancing Through Soundscapes: Designing a Low-Cost Sound-Based Device for Sensing and Interpreting Movement and Dance
链接: https://arxiv.org/abs/2608.19827
作者: Swen E. Gaudl,Silvia Carderelli-Gronau
类目: Human-Computer Interaction (cs.HC); Sound (cs.SD)
备注: full paper on technical artifact for TEI2027 conference
Abstract:When we move through space, we often rely on multiple senses beyond vision to perceive and act in that environment: we ``feel’’ the presence of others; we build internal representations and models and recall them to navigate the environment. We also leave traces and impressions that others pick up on. The traces include echoes, heat, the displacement of objects such as furniture or footprints, air movement close to the face of another, smells such as perfume, but also the immediate sounds we make when we move and breathe. Movement is a spatial and temporal activity, and dance as a form of movement practice requires coordination of oneself in relation to others, the space and a potential score. When rehearsing dance, dancers have to relate to others often not just by looking but more often by feeling and imagining or remembering where others are based on experience and shared practice. So how can we approach technology-mediated movement and dance? Why should we explore it? How can we support spatial and temporal practice meaningfully and joyfully? In this work, we focus on sound traces; we present the design and rationale for a sound-based artefact that translates movement-based sound into layered, explorable, generative soundscapes. The work contributes a novel artefact for exploring movement-based activities with a audio-first approach, with a focus on the spatial performative experience. The paper further reflects on observations from workshops and public sharings, including how participants used repetition, stillness, environmental sound, and call-and-response to understand and improvise with the soundscape.
[HC-6] Calming Robot Pitches? Exploring the Influence of Robot Voice Pitch on Childrens Stress Levels
链接: https://arxiv.org/abs/2608.19826
作者: Nina G. M. van Roij,Emilia I. Barakova,Briana Isaila,Aoju Chen
类目: Human-Computer Interaction (cs.HC); Robotics (cs.RO)
备注:
Abstract:This study examined whether variations in robot speech pitch influence children’s stress levels during a robot-guided game. Although lower-pitched voices have been shown to facilitate stress regulation in human communication, it remains unclear whether this effect generalizes to synthetic voices in child-robot interactions. Twenty-seven Dutch children aged 8-12 years were randomly assigned to interact with a Zenbo Junior II robot using either a lower-pitched or a higher-pitched voice. The interaction consisted of an introduction followed by a timed LEGO-building game. Stress levels, measured with an adapted version of CAM-S, increased during the game, confirming the stress-inducing nature of the task. No differences emerged between pitch conditions. These findings suggest that the benefits of lower pitch in reducing stress may not directly translate to child-robot interactions. Possible explanations include children’s developing sensitivity to emotional tone, mismatches between the robot’s voice and appearance, or the use of fixed pitch changes that sound unnatural, since real speech varies dynamically across multiple dimensions. Future research examining combinations of prosodic cues (beyond pitch alone) could provide further insights and help inform robot voice design for effective stress regulation support for children.
[HC-7] Grounding Mindfulness in Embodied Tangibles: A Scoping Review Theoretical Framework for HCI Design
链接: https://arxiv.org/abs/2608.19673
作者: Tharaka Sachintha Ratnayake,Samangi Wadinambiarachchi,Sarah Schömbs,Jonathan Eden,Denny Oetomo,Wafa Johal
类目: Human-Computer Interaction (cs.HC)
备注:
Abstract:Embodied and tangible devices are increasingly used to support mindfulness practices across meditation, yoga, and everyday routines. However, existing HCI research lacks a coherent theoretical foundation for explaining how such systems support distinct mindfulness processes and outcomes. First, we report findings from a scoping review of tangible devices (n=65) for mindfulness in HCI based on the mechanisms of action of mindfulness. The review found that most systems primarily target attentional regulation and body awareness, while emotion regulation and change in perspective on the self remain comparatively underexplored. Also, the evaluation methods used to assess the effectiveness of tangible systems for mindfulness-related outcomes were found to be fragmented and weakly grounded in theory. Building on these findings, a theoretical framework grounded in the Self-Awareness, Self-Regulation, and Self-Transcendence (S-ART) framework is proposed to explain how tangible systems support different forms of mindfulness practice. Within the framework, we present two complementary models: (1) Embodied Sensory Expansion (ESE) for focused-attention meditation (FAM) and (2) Embodied Sensory Anchoring (ESA) for open-monitoring meditation (OMM). Together, we show how this framework can provide a principled basis for explaining designs, generating hypotheses, and evaluating tangible mindfulness technologies in HCI, while highlighting key gaps for future research.
[HC-8] IRIS: Navigating and Reflecting on Writing Traces Using Intelligent Document Histories
链接: https://arxiv.org/abs/2608.19614
作者: David Zhou,Andrew Chen,John Joon Young Chung,Sarah Sterman
类目: Human-Computer Interaction (cs.HC)
备注: 18 pages, 7 figures, ACM Symposium on User Interface Software and Technology (UIST) 2026
Abstract:Much of the text produced throughout the lifetime of a document is impermanent. In this paper, we explore how writing activity traces can be made visible and interactive to help writers navigate their document histories and understand their writing processes. Using the Flower and Hayes cognitive process model of writing, IRIS infers writing process states from keystroke logs and presents them using an AI-enhanced version history. IRIS provides three primary interactions: revision highlighting that shows local process histories in-situ, conceptual filters that constrain the version history by process type or topic, and natural language inquiry that lets writers pose reflective questions about their writing and process. Following a formative and a longitudinal study, we find that writers use the interfaces to locate specific revisions and understand the progression of their writing. They use system outputs as interpretive material, relating them to pre-existing beliefs and confirming, challenging, and deepening their understanding of their writing.
[HC-9] Localized Ecological Momentary Assessment for Mental Health Research in China: An Implementation-Oriented Framework and Preliminary Case Application
链接: https://arxiv.org/abs/2608.19588
作者: Xinying Zhao,Yue Li,Jiafeng Wang,Yunfan Fu,Ruilin Guo,Chen Yang,Cheng Yao,Wei Deng
类目: Human-Computer Interaction (cs.HC)
备注:
Abstract:Background: Ecological momentary assessment (EMA) is increasingly used in mental health research, but research-grade deployment requires platforms supporting protocol configuration, automated delivery, participant management, and data export. In China, these requirements are not consistently supported. Objective: We aimed to identify workflow gaps affecting localized EMA deployment, develop an implementation-oriented framework for platform assessment, and assess Huixin EMAI. Methods: We reviewed EMA platforms reported in Chinese mental health studies in CNKI and Wanfang. A multidisciplinary panel of 6 experts developed the Multi-dimensional EMA Platform Evaluation Framework (MEPEF) and benchmarked 7 platforms across 43 indicators in 6 domains. MEPEF was then applied to Huixin EMAI using deployment logs from 48 participants, questionnaires from 44 participants, and semistructured interviews with 6 researchers. Results: We identified 66 empirical studies. Most relied on instant-messaging-based workflows (36/66, 54.5%), whereas specialized EMA platforms were less common (14/66, 21.2%). MEPEF provided a 6-domain framework for cross-platform benchmarking and highlighted a trade-off between localized deployability and advanced research functions. In a Huixin EMAI deployment, 1893 of 2472 expected prompts were completed (76.6%), with a median response latency of 4.0 minutes (Q1-Q3 0.0-13.0). Participant feedback indicated favorable acceptability; researchers reported support for core workflows but gaps in control, delivery monitoring, and data readiness. Conclusions: The main challenge for EMA in Chinese mental health research appears to lie less in feasibility than in recurring workflow gaps affecting localized deployment. This study translates these gaps into structured evaluation and design targets, providing an implementation-oriented pathway for advancing localized EMA platforms.
[HC-10] Delegating or Doing? Understanding User Behavior in Hybrid Human-Agent Interfaces
链接: https://arxiv.org/abs/2608.19551
作者: Gavin Raine Dizon,Tyrone Justin Sta Maria,Jordan Aiko Deja,Yasuyuki Sumi
类目: Human-Computer Interaction (cs.HC)
备注: 9 pages, In Proceedings of the 14th International Conference on Human-Agent Interaction at Osaka, Japan on November 16-19, 2026
Abstract:Large Language Models (LLMs) are increasingly embedded into applications, allowing users to complete tasks either through direct manipulation or by delegating actions to conversational agents. However, little is known about how users balance these modalities when both are available. We present a web-based content management system augmented with an LLM agent through the Model Context Protocol (MCP), enabling users to perform CRUD tasks through a graphical interface, a conversational agent, or both. We conducted a between-subjects study (N=73) comparing three interaction modes: Traditional-Only, AI-First, and Hybrid. Across sixteen scenarios, we analyzed task completion time, interaction logs, and delegation behavior. AI-assisted interaction significantly reduced clicks, page navigations, and scrolling indicating lower interaction effort. Surprisingly, these reductions did not translate into faster task completion, as task duration did not differ significantly across conditions. We also found no significant relationship between CRUD operation type and delegation, suggesting that users did not systematically avoid delegating higher-risk actions. Instead, delegation varied far more between participants than between tasks, with individual differences accounting for roughly half the variance in assistant use (ICC = .50). Our findings suggest that the primary benefit of human–agent interfaces may be reducing interaction effort rather than improving speed, and that delegation reflects who the user is more than what the task demands.
[HC-11] Generating Diverse Personas for User Simulators to Test Interview Dialogue Systems SIGDIAL2025
链接: https://arxiv.org/abs/2608.19549
作者: Mikio Nakano,Kazunori Komatani,Hironori Takeuchi
类目: Computation and Language (cs.CL); Human-Computer Interaction (cs.HC)
备注: Accepted for publication at SIGDIAL 2025, 19 pages, 18 figures,
Abstract:This paper addresses the issue of the significant labor required to test interview dialogue systems. While interview dialogue systems are expected to be useful in various scenarios, like other dialogue systems, testing them with human users requires significant effort and cost. Therefore, testing with user simulators can be beneficial. Since most conventional user simulators have been primarily designed for training task-oriented dialogue systems, little attention has been paid to the personas of the simulated users. During development, testing interview dialogue systems requires simulating a wide range of user behaviors, but manually creating a large number of personas is labor-intensive. We propose a method that automatically generates personas for user simulators using a large language model. Furthermore, by assigning personality traits related to communication styles when generating personas, we aim to increase the diversity of communication styles in the user simulator. Experimental results show that the proposed method enables the user simulator to generate utterances with greater variation.
[HC-12] Does Listening Matter? Backchanneling and Nodding in AI Clone
链接: https://arxiv.org/abs/2608.19527
作者: Koji Inoue,Kazushi Kato,Tatsuya Kawahara,Shunichi Kasahara
类目: Human-Computer Interaction (cs.HC); Computation and Language (cs.CL); Sound (cs.SD)
备注: This paper has been accepted to the Late-Breaking Results (LBR) track of the 28th International Conference on Multimodal Interaction (ICMI 2026)
Abstract:AI clones that imitate a specific person typically reproduce what the person says and how they sound, but not how they listen. We investigate whether adding multimodal listening behaviors gives such a clone more presence and authenticity. We integrated verbal backchannels and head nodding, driven by real-time prediction models, into an AI clone equipped with voice cloning and LLM-based responses. In a within-subjects study (N=35), adding these behaviors significantly improved the perceived attentiveness of the avatar, the sense of talking with the real person, and the feeling of co-presence. These results indicate that AI clone fidelity should extend beyond voice and response content to include interactive listening behavior.
[HC-13] Social.Wiki: A Web Held in Common
链接: https://arxiv.org/abs/2608.19433
作者: Theia Henderson,Carmel Schare,Ana Dodik,Clemens N. Klokmose,Ziv Epstein,David D. Clark,David R. Karger
类目: ocial and Information Networks (cs.SI); Computers and Society (cs.CY); Human-Computer Interaction (cs.HC)
备注: Accepted to The 39th Annual ACM Symposium on User Interface Software and Technology (UIST '26), November 02-05, 2026, Detroit, MI, USA
Abstract:Many of the websites people depend on have owners whose interests are not fully aligned with their users. We address the root of this problem by presenting a reimagining of the web where sites are not owned at all but are instead collaboratively produced like Wikipedia articles. We call the system this http URL because it supports the co-creation of interactive social sites, such as those for microblogging, messaging, dating, gaming, ride sharing, and so on. With off-the-shelf AI tools, people with little or no programming experience can edit these sites to better reflect the needs and preferences of their communities. this http URL builds on ideas from collaborative malleable software systems such as Webstrates, but is designed for public participation rather than use only within small, trusted groups. To this end, this http URL includes governance to mitigate conflict. To accommodate diverse governance preferences, our model of “plural governance” lets people independently choose the policies that determine which edits to a site they see. this http URL also implements a granular security model to protect personal data in a malleable environment. Complementing the decentralized design and governance of this http URL sites, both site edits and within-site data are stored on Graffiti, a decentralized infrastructure, decoupling the ownership of underlying servers from the ownership of sites. We evaluate this http URL through case studies that demonstrate the range of sociotechnical structures it supports, as well as through deployments at a hackathon and in the wild. Comments: Accepted to The 39th Annual ACM Symposium on User Interface Software and Technology (UIST '26), November 02-05, 2026, Detroit, MI, USA Subjects: Social and Information Networks (cs.SI); Computers and Society (cs.CY); Human-Computer Interaction (cs.HC) Cite as: arXiv:2608.19433 [cs.SI] (or arXiv:2608.19433v1 [cs.SI] for this version) https://doi.org/10.48550/arXiv.2608.19433 Focus to learn more arXiv-issued DOI via DataCite (pending registration) Related DOI: https://doi.org/10.1145/3830398.3830639 Focus to learn more DOI(s) linking to related resources Submission history From: Theia Henderson [view email] [v1] Wed, 19 Aug 2026 20:31:16 UTC (8,504 KB)
[HC-14] Cyber-Physical Systems for Accessibility and Ability Augmentation: Bridging Diverse Communities
链接: https://arxiv.org/abs/2608.19422
作者: Shuchang Xu,Riku Arakawa,Mina Huh,Nandi Zhang,Tianyu Zhang,Wazeer Zulfikar,Ruei-Che Chang,Yotam Sechayk,Huamin Qu,Amy Pavel,Franklin Mingzhe Li,Yukang Yan,Brian A. Smith,Pattie Maes
类目: Human-Computer Interaction (cs.HC)
备注: UIST 26 Workshop
Abstract:The powerful convergence of wearables, robotics, extended reality, and smart environments is expanding the design space for cyber-physical systems (CPS) that support and augment human abilities in daily life. By sensing real-world contexts, modeling user needs, and providing situated assistance, these systems can improve accessibility for people with disabilities while enhancing broader human abilities such as perception, memory, learning, and mobility. However, realizing this potential requires addressing key challenges in context sensing, user modeling, adaptive interaction, privacy, and evaluation to ensure that CPS are reliable and effective in real-world contexts. This workshop will bring together researchers and practitioners across HCI, AI, wearables, robotics, XR, smart environments, accessibility, and ability augmentation to examine shared strategies and challenges for designing accessibility- and ability-centered CPS. Through panel discussions, interactive demos, and mixed-group design activities, participants will identify recurring design principles, technical challenges, and future directions for CPS that support and augment human abilities in real-world settings. For details, please visit: this https URL.
[HC-15] Scientific Visualization as a Collaborative Data Infrastructure
链接: https://arxiv.org/abs/2608.19413
作者: Jasmine Tan Otto,Alexandra Diehl,Max Kreminski,Scott Davidoff
类目: Human-Computer Interaction (cs.HC)
备注: to appear in CSCW 2026 Workshop on Infrastructure and Collaborative Ecologies
Abstract:Scientific visualization is an active site of infrastructuring with many layers of data and evidential claims. This paper reflects on collaborative Mars geoscience research conducted at the NASA Jet Propulsion Lab, which produced the PIXLISE spectroscopic analysis platform, through a retrospective analysis of a scientific discovery made by the team using PIXLISE. The history of infrastructuring in scientific visualization is exceedingly rich, which presents an opportunity for archival research. Simultaneously, the field of visualization aspires to become a science of communication, opening the door to future collaborations. Finally, with regard to visualization practitioners, we believe that opportunities for infrastructuring in both science and science communication are actively emerging.
[HC-16] Multi-Tier Mentorship with AI-Assisted Development: Authentic Engineering for K-12 and Undergraduates
链接: https://arxiv.org/abs/2608.19379
作者: Kelly Yuan,Ronald Liu,Daniel Crawford,Weihao Qu
类目: Computers and Society (cs.CY); Human-Computer Interaction (cs.HC)
备注: 8 pages, 8 figures, 2 tables. Accepted to IEEE ISEC 2026
Abstract:K-12 students often possess creative engineering ideas but lack technical skills to build them, while undergraduates have coding expertise but few opportunities to lead real-world projects or mentor others. The rapid development of AI-assisted tools offers a potential bridge to connect these groups, yet the structure for effective K-12 and university collaborations remains underexplored. This paper introduces a multi-tiered mentorship framework enabling high school students to engage in authentic engineering through AI-assisted development using large language models and AI agents, while undergraduate mentors provide architectural oversight. We test this framework through LuckyTag, a privacy-preserving NFC-based lost-and-found system. The model positions high schoolers as product leads, undergraduates as technical architects, and faculty as minimal-intervention advisors. A pilot with four high school students, three undergraduates and two faculty yielded survey data showing high perceived barrier removal and gains in system architecture understanding. Thematic analysis reveals that AI amplifies rather than supplants mentoring demands, requiring human oversight for logic and security. These findings suggest a hybrid model for equitable K-12 and university collaboration on computing integration that emphasizes “AI micromanagement” and architectural reasoning over traditional syntax.
[HC-17] MultiVerse: A Creator-Centered Approach to Steering Context-Adaptive Lyrics
链接: https://arxiv.org/abs/2608.19350
作者: Alexander Wang,Chris Donahue,David Lindlbauer
类目: Human-Computer Interaction (cs.HC); Sound (cs.SD)
备注:
Abstract:Generative AI may enable new forms of context-aware creative expression by dynamically tailoring media content to its consumption context. For instance, AI systems could adapt song lyrics to the listener and their current activity. However, existing media adaptation systems primarily optimize for audience experience, often neglecting artists’ intent, style, and preference. We address this challenge by introducing a novel creator-centered approach to adaptive media authoring and present MultiVerse, a system that instantiates this approach for steering adaptive lyrics. Our approach allows creators to explicitly author controls based on their intent, lyric structure, and audience context, and uses rule-based validations to ensure controls are followed. We conducted a study with 10 songwriters, comparing MultiVerse with a prompting-based workflow for composing adaptive lyrics. The comparison revealed that creators preferred to author how lyrics adapt by explicitly specifying relevant context and adaptation constraints, while recognizing trade-offs in flexibility and iteration speed. Interviews further revealed that creators viewed adaptive media as enabling new forms of audience connection, introducing a distinct creative process, favoring new compositional strategies, and reshaping notions of authorship.
[HC-18] APPROVE: Visual End-User-in-the-Loop Robot Programming with LLM s ICME2026
链接: https://arxiv.org/abs/2608.19281
作者: Bijan Kavousian,Miray Özakkas,Josefine Monnet,Oliver Petrovic,Christian Brecher
类目: Robotics (cs.RO); Human-Computer Interaction (cs.HC)
备注: Accepted for publication in Procedia CIRP, Proceedings of the 20th CIRP Conference on Intelligent Computation in Manufacturing Engineering (ICME 2026)
Abstract:Programming robots remains challenging for non-experts, as traditional methods require expert knowledge and even block-based interfaces often lack flexibility. Recent work has explored Large Language Models (LLMs) to automatically generate robot programs from natural language, but these systems remain limited by a lack of transparency, missing mechanisms to ensure alignment with user intent, and little support for reuse. We present APPROVE (AI-Powered Programming for Robots with Visual End-User Feedback), an LLM-based multi-modal end-user programming framework that integrates natural language input with a block-based interface and an explicit user confirmation step. Generated programs are visualized using a block-based interface in Blockly, allowing users to confirm, modify, or reject them before execution. Confirmed functions are stored in a library for reuse, gradually building a set of reliable program components. Our approach contributes a human-centered design for LLM-based robot programming that emphasizes user trust, intent alignment, and reusability.
计算机视觉
[CV-0] WithEveryone: Unified Planning and Identity Grounding for Group Image Generation
链接: https://arxiv.org/abs/2608.20336
作者: Hengyuan Xu,Qixun Wang,Yiji Cheng,Miles Yang,Zhao Zhong,Wei Cheng,Xingjun Ma,Yu-gang Jiang
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: Project Page: this http URL ;Code will be released: this http URL
Abstract:Identity-preserving image generation becomes increasingly unreliable when a scene must contain many specified people. Beyond retaining each identity, the model must bind every reference to a distinct person and location, while training-time identity losses must establish correspondence among several noisy predicted faces. We introduce WithEveryone, a unified framework for generating group images up to ten reference identities. WithEveryone injects each selected identity as an addressed token, predicts a structured identity–layout plan, and renders the plan as a visual condition. Its key objective, Layout-Grounded ID Loss, uses annotated face regions to supervise the intended identities directly, avoiding unstable embedding-based face matching; ID Representation Forcing additionally trains a prediction for each identity before image synthesis. On an identity-disjoint benchmark, WithEveryone achieves the highest target-context identity similarity, improving face similarity from 0.462 for GPT-Image-2 to 0.499, while reducing copy-paste artifacts from 0.169 to 0.055. It further covers 97.3% of the requested identities with a duplicate rate of only 2.8%. These results show that explicit identity–layout grounding enables identity-preserving generation to scale to larger groups without relying on direct reference-face copying.
[CV-1] 4DAnyone: Create Anyone in 4D from a Casual Monocular Video
链接: https://arxiv.org/abs/2608.20335
作者: Yudong Jin,Tao Xie,Qihang Zhang,Zehong Shen,Zhen Xu,Yujun Shen,Hujun Bao,Xiaowei Zhou,Yinghao Xu
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: Project page: this https URL
Abstract:We present 4DAnyone, a framework for reconstructing 4D humans from an uncalibrated monocular video by generating reconstruction-grade multiview-consistent videos and lifting them into 4D Gaussian Splatting (4DGS). Existing camera-controlled video diffusion models synthesize plausible novel-view videos but fail to maintain consistency when scaled to the tens of target views required for 4DGS reconstruction. We identify this failure as a bounded-attention-context problem: when target views exceed the capacity of a single DiT forward pass, they must be split into groups, exposing two coupled bottlenecks. On the reference-context side, conditioning on all previously generated views grows as O(N) , weakening cross-view appearance guidance. On the target-context side, disjoint groups cannot directly exchange information, causing global structural drift. 4DAnyone addresses both bottlenecks with two complementary designs: Reference Context Packing (RCP) compresses growing reference views into a fixed-length mixed-resolution context with O(1) reference-context complexity, while Target Context Routing (TCR) rotates target-view groupings during denoising to share context across groups at high-noise steps and stabilize details at low-noise steps. We further build the MVGameHuman dataset using our in-house game engine and combine it with light-stage and in-the-wild video datasets for training. Experiments on DNA-Rendering and DyMVHumans show that 4DAnyone outperforms prior methods in both novel-view video quality and downstream 4DGS reconstruction, with robust in-the-wild generalization. See our project page for video results and source code: this https URL.
[CV-2] Swift-Image: Exploring the Performance Frontier of Compact Unified Image Generation Models
链接: https://arxiv.org/abs/2608.20334
作者: Taihang Hu,Zhao Wang,Zuan Gao,Tao Liu,Hao Yan,Zhengze Xu,Yuhang Yu,Yongchao Du,Xingjian Wang,Jun Zheng,Qinye Zhou,Zhengrui Chen,Chao Lin,Yefeng Shen,Zhengtao Wu,Ge Wu,Xiaoli Xu,Denghui Yang,Huayu Zhang,Mingzhou Zhang,Mengting Chen
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: 28 pages, 11 figures
Abstract:We present Swift-Image, a compact unified model for text-to-image generation, single-image editing, and multi-image editing. Our goal is to explore how far a relatively small visual generator can be pushed through systematic training engineering under a constrained computational budget. Swift-Image adopts an efficient 6B single-stream DiT and a progressive training pipeline that evolves from broad semantic coverage to higher resolution, stronger visual quality, and unified generation-editing supervision. For post-training, we employ parallel expert reinforcement learning followed by multi-teacher on-policy distillation to alleviate interference among heterogeneous objectives. We further decouple high-level reasoning from pixel-level rendering with a Prompt Enhancer that translates user requests into generator-aligned visual specifications. For efficient deployment, structural pruning and few-step distillation produce 3B and accelerated variants. Swift-Image achieves leading aggregate performance among evaluated open-source models with only 6B parameters and 243K GPU training hours; the compressed 3B model incurs nearly no loss, while few-step distillation further improves aggregate editing performance with substantially fewer sampling steps. Our study also summarizes practical lessons for architecture, data curriculum, post-training, prompt enhancement, and model compression.
[CV-3] Inter-X: A Comprehensive Benchmark for Multimodal Human-Human Interaction Analysis
链接: https://arxiv.org/abs/2608.20312
作者: Liang Xu,Chengqun Yang,Zili Lin,Xintao Lv,Yichao Yan,Xin Jin,Zhibo Chen,Xiaokang Yang,Wenjun Zeng
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: 24 pages, 10 figures
Abstract:The capability to perceive and synthesize human-human interactions is fundamental to developing intelligent digital human systems. However, existing datasets and modeling approaches are fundamentally constrained by low-fidelity kinematics, the omission of dexterous hand gestures and a severe lack of rich multimodal annotations. Furthermore, fragmented interaction representations and inconsistent evaluation protocols also impede fair and rigorous benchmarking. To systematically address these bottlenecks, we present Inter-X++, a comprehensive and large-scale benchmark designed to empower versatile HHI analysis. Captured via a novel hybrid motion capture system, Inter-X++ provides 11,388 high-fidelity interaction sequences and over 8.1M frames, featuring precise whole-body movements and detailed finger articulations. Meanwhile, we enrich the data foundation with multifaceted annotations, including hierarchical fine-grained textual descriptions, interaction categories, causal interaction orders, the relationship and personality of the subjects, as well as vertex-level contact maps and physically regularized constraints. Leveraging these elaborate annotations, we formulate a unified testing ground comprising four categories of downstream tasks that symmetrically span both generative and perceptive paradigms. To eliminate benchmarking ambiguities, we systematically standardize the interaction representations and evaluation protocols. Finally, we go beyond dataset construction to propose OpenHHI, a single and unified HHI representation and modeling framework that jointly optimizes interaction reconstruction and semantic understanding. Extensive experiments reveal that OpenHHI achieves state-of-the-art performance on both generation and perception tasks. This definitively proves that our unified representation successfully bridges interaction understanding and generation simultaneously.
[CV-4] DreamHand: Repurposing Video Diffusion Models for Occlusion-Robust Egocentric 3D Hand Motion Recovery
链接: https://arxiv.org/abs/2608.20308
作者: Yufei Liu,Xixi Wang,Hao Li,Ganlong Zhao,Kaitong Cai,Chengkai Jin,Chunxiao Liu,Jianbo Liu,Siyuan Huang,Xingang Pan,Hongsheng Li
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: Project Page: this https URL
Abstract:Egocentric video offers scalable manipulation data for embodied AI, yet recovering metric 3D hand trajectories remains challenging due to severe object occlusion and frequent out-of-sight gaps. Existing single-frame and windowed temporal regressors fail when hand shortly leaves the frame, while recent video diffusion models (VDMs) rely on heavy, stochastic multi-step sampling as pixel-space renderers. We instead repurpose VDM into a deterministic geometry encoder. A single forward pass over the clean latent exposes scene content beyond current observations, including occluded and out-of-sight hands. We introduce DreamHand, an offline clip-level framework that extracts features via a Deterministic Clean-Latent Encoder and decodes them with a Bidirectional Spatiotemporal Decoder. DreamHand recovers continuous bimanual trajectories with metric placement and no external detector, while a Ray-Based Camera Solver supports a second configuration that needs no test-time camera intrinsics. Across five egocentric benchmarks, DreamHand sets a new state of the art, cutting MPJPE-p by 30% on occlusion-heavy ARCTIC and 40% on HOT3D. These gains reach 46%-61% once out-of-sight hands are included in the evaluation, offering a scalable path from everyday human video to robot manipulation data.
[CV-5] CalcSeg: Confidence-aware 3D Latent Context Curriculum Learning For Myocardial Scar Segmentation From Single-Stack LGE-CMRs
链接: https://arxiv.org/abs/2608.20305
作者: Nivetha Jayakumar,Hannah Kim,Amit R. Patel,Miaomiao Zhang
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Myocardial scar segmentation from single-stack late gadolinium-enhanced cardiac magnetic resonance (LGE-CMR) imaging has been a longstanding and clinically important challenge, particularly in the presence of low tissue contrast, diffuse, and small scar regions. These challenges are further intensified by the limited availability of 3D spatial context. This paper presents CalcSeg, a Confidence-aware latent context curriculum learning framework that leverages fused 3D feature representations from single-stack 2D LGE-CMR images for robust scar segmentation. Specifically, we introduce a dynamic semi-supervised curriculum learning strategy that progressively expands training from easier to more challenging scar cases using a learned confidence-aware scoring function. Such a function integrates errors in the predicted scar maps with quantified epistemic uncertainty and scar burden estimation to automatically assess sample difficulty without requiring manual labels. To compensate for the limited spatial context in single-stack acquisitions, we then develop a latent slice-wise self-attention to capture inter-slice dependencies and infer 3D spatial representations from sparse 2D inputs. We evaluate CalcSeg on multi-center clinical LGE-CMR datasets and benchmark against existing scar segmentation networks. Experimental results show that CalcSeg consistently outperforms all competing methods, particularly with substantial improvements on clinically challenging cases. Our code is released on Github.
[CV-6] owards Surgical World-Action Modeling: A Preliminary Joint Visual-Trajectory Forecasting for Surgical Motion Planning
链接: https://arxiv.org/abs/2608.20284
作者: Weiliang Huang,Huanrong Liu,Bob Zhang,Qi Dou,Zhen Chen,Yun Gu,Guy Rosman,Qingbiao Li
类目: Computer Vision and Pattern Recognition (cs.CV); Robotics (cs.RO)
备注:
Abstract:Reliable surgical planning requires models to anticipate not only how instruments will move, but also how the operative visual state will evolve together with such motion. Existing approaches typically treat future scene generation and instrument trajectory prediction as two separate tasks. Scene-only models cannot directly evaluate the accuracy of future instrument motion at the trajectory level, while trajectory-only models fail to capture the visual consequences of instrument movement, leaving the consistency between predicted trajectories and future scene evolution unaddressed. Jointly forecasting both provides a more complete account of surgical action-scene dynamics by enabling explicit trajectory-level evaluation while simultaneously modeling the corresponding visual evolution. To bridge this gap, we present a preliminary joint visual-trajectory world-action model that simultaneously forecasts future visual states and instrument trajectories from historical surgical observations. Specifically, we encode historical video frames and tool trajectories into latent representations, which are processed by a temporal-spatial encoder and subsequently decoded through separate visual-state and trajectory prediction heads. Based on this preliminary architecture, a chunked autoregressive rollout is repeatedly applied to predict fifteen future steps. The chunked strategy consistently outperforms direct one-shot prediction across all evaluated horizons, improving first-segment PSNR from 18.86 to 23.11 dB and reducing ADE from 45.77 to 22.22 pixels. These results demonstrate the initial feasibility of joint visual-motion forecasting. However, we observe progressive visual degradation and accumulated trajectory errors over longer prediction horizons, which remain important challenges for future surgical world-action modeling.
[CV-7] Ultra-High-Definition Restoration Transformers with Correlation Matching Transformation
链接: https://arxiv.org/abs/2608.20263
作者: Cong Wang,Liyan Wang,Jinshan Pan,Wei Wang,Wenqi Ren,Jun Liu,Xiaochun Cao
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:We propose UHDformer++, a general Transformer-based framework to solve numerous Ultra-High-Definition (UHD) image restoration tasks. UHDformer++ operates across 4 coordinated learning spaces: 1) a high-resolution space (HR) for multi-level feature extraction, 2) a low-resolution space (LR) for learning compact, representative features, 3) a super-resolution space (SR) for upsampling low-resolution features from SR, and 4) a low-high fusion and reconstruction space (LHFR) for final image restoration. Specifically, HR extracts multi-scale high-resolution features and fuses them with low-resolution cues to produce residual images, while LR distills complementary representations from HR to improve restoration quality. To supply LHFR with richer features, SR super-resolves LR outputs before fusion. We further introduce two modules to bridge the high- and low-resolution spaces. The Feature-Refined Correlation Matching Transformation (FR-CMT) module selects the top C/r~(C~\textdenotes the number of channels;~r\geq1~\textcontrols the squeezing level) , from the fusion between max- and mean-pooled high-resolution features to replace less informative channels in the low-resolution Transformer. The Adaptive Channel Modulator (ACM) adaptively recalibrates multi-scale high-resolution features, ensuring that only task-relevant information propagates to LR. Extensive experiments demonstrate that UHDformer++ reduces model parameters by at least 86% compared with recent state-of-the-art methods while achieving substantial performance gains across 5 UHD restoration tasks, including low-light image enhancement, dehazing, deblurring, deraining, and desnowing. Code will be released at this https URL.
[CV-8] Prompt-Conditioned Channel Attention for Hierarchical Feature Modulation toward Anatomy-Agnostic Segmentation
链接: https://arxiv.org/abs/2608.20229
作者: Mosharof Hossain,Md Rabiul Islam,Limon Halder,Erchin Serpedin,Md Kamrul Hasan
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
备注:
Abstract:Anatomically plausible segmentation remains challenging because of low contrast, ambiguous boundaries, and modality-specific artifacts. Interactive segmentation has emerged as a promising strategy to guide feature extraction and improve localization, particularly in structurally ambiguous regions. However, existing methods integrate prompts through late-stage fusion and lack explicit mechanisms for prompt-driven channel-wise modulation across hierarchical feature representations, limiting their ability to capture deeper contextual and modality-specific variations. To address these limitations, we introduce Prompt-Conditioned Channel Attention (PCCA), a novel modulation mechanism that enables deep, hierarchical integration of semantic prompts within encoder-decoder networks. PCCA extracts compact channel descriptors via pooling, projects them into a shared space, and fuses them through a gated excitation mechanism to compute prompt-aware channel attention weights. These weights adaptively recalibrate feature responses across multiple network stages, enabling prompt-conditioned, semantically enriched hierarchical representations. Building on this, we propose PROMISE-Net, instantiated in two network variants: a convolutional model (PROMISE-CNN) and a transformer-based model (PROMISE-Txformer). Across the ISIC-Lesion, Kvasir-Polyp, CAMUS-Cardiac, and Kvasir-Instrument benchmarks, integrating PCCA into PROMISE-CNN yielded relative IoU gains of 10.4%, 8.7%, 0.8%, and 3.4%, respectively, over the baseline U-Net, while PROMISE-Txformer achieved corresponding gains of 7.6%, 23.0%, 2.1%, and 1.1%, respectively, over the baseline UNETR. These results show consistent improvements across architectures, imaging modalities, and anatomical targets, establishing PCCA and PROMISE-Net as a scalable, generalizable framework for prompt-aware hierarchical feature modulation in medical image segmentation.
[CV-9] Unwarping the Lens: A Physics-Grounded Approach to Video Glasses Removal
链接: https://arxiv.org/abs/2608.20212
作者: Radim Spetlik,David Futschik,Radek Danecek,Feitong Tan,Ziqian Bai,Rohit Pandey,Yinda Zhang
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:High-fidelity removal of eyeglasses from video is a major challenge in facial attribute editing, as the underlying facial geometry is often obscured by complex refractive distortions and view-dependent specular reflections. While large-scale generative priors have shown promise in eye-glasses removal via static image inpainting, they often lack the structural constraints necessary to maintain identity, expression, and pose, leading to visible “identity drift” in both static images and dynamic sequences. In this paper, we propose a novel transfer framework that addresses the stochastic nature of generative priors. Our pipeline first extracts high-fidelity synthetic face images from a commercial-grade generative model (Nano Banana, Gemini 3 Pro Image), regularizes them via a three-stage structural filtering process to preserve identity, expression, and pose, and finally applies physically-based simulation of lens optics during training to provide diverse, paired data. This process transfers Nano Banana’s photo-realistic, multi-view knowledge into a specialized restoration architecture, JFSnet (Joint Feature-Spatial network). JFSnet integrates DINOv2-based semantic features with a convolutional decoder for spatial reconstruction, leveraging translation equivariance constraints to improve temporal consistency and high-frequency detail preservation. Evaluations on the curated Flickr-Faces-HQ (FFHQ) subset (12,163 images) show that our approach achieves high fidelity and structural accuracy, while maintaining inference speed of 27.68 FPS. In perceptual studies on CelebV-Text video sequences, our results are consistently preferred over diffusion and GAN-based baselines for ocular consistency, temporal stability, and overall restoration quality.
[CV-10] RoMAN-Flow: Taming Autoregressive Normalizing Flows for Offline Reinforcement Learning in Robotic Manipulation
链接: https://arxiv.org/abs/2608.20208
作者: Shaoxuan Wang,Guangting Zheng,Rui Huang,Zhipeng Tang,Sha Zhang,Jiajun Deng,Yanyong Zhang
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Offline reinforcement learning improves robotic policies using previously collected data without further environment interaction. Yet prevalent diffusion- and flow-matching robot policies lack tractable likelihoods, limiting their use in likelihood-based offline RL post-training. AR-NFs offer both expressive action modeling and exact likelihood evaluation, but their sequential sampling incurs substantial sampling overhead during policy optimization and deployment. We present RoMAN-Flow (Robotic Manipulation with Autoregressive Normalizing Flows), an offline reinforcement learning framework that makes AR-NF policies practical for robotic manipulation by addressing this sampling bottleneck in both stages. During policy optimization, RoMAN-Flow employs a sampling-free, advantage-weighted likelihood objective that assigns higher likelihood to high-advantage actions from the offline dataset without sampling from the autoregressive policy. For efficient deployment, it distills the optimized autoregressive policy into a one-step action generator, enabling low-latency action prediction. Experiments across multiple simulated manipulation benchmarks and real-world robotic platforms demonstrate that RoMAN-Flow achieves competitive policy performance while substantially reducing inference latency. Code is available at this https URL.
[CV-11] G3Ego: Gaze-Guided Graphs for Egocentric Action Understanding ECCV2026
链接: https://arxiv.org/abs/2608.20157
作者: Marko Haralović,Akash Ramakrishnan,Estefania Talavera Martinez
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: Accepted at the CONTEXTUS Workshop, ECCV 2026
Abstract:Egocentric action understanding is often addressed using large video models pretrained on extensive exocentric datasets. However, many first-person actions depend on a small number of hand-object interactions involving only a few relevant entities. We propose G3Ego, a graph-based framework for egocentric action understanding that uses gaze as a structural cue to identify action-relevant entities in the scene. From sparsely sampled frames, G3Ego constructs action scene graphs from vision-language descriptions, grounded objects, and hand cues, and then prunes irrelevant entities using the camera wearer’s gaze. The resulting graph embeddings are temporally aggregated for action recognition and anticipation. Unlike prior work that uses gaze primarily as an auxiliary modality or attention signal, G3Ego incorporates gaze directly into graph construction, producing efficient and interpretable representations focused on action-relevant interactions. Experiments on EGTEA Gaze+ and MECCANO show that G3Ego achieves competitive performance compared with video-based approaches and consistently improves Macro-F1 under class-imbalanced evaluation, while avoiding reliance on computationally expensive video pretraining. These results demonstrate the effectiveness of gaze-guided graph representations for egocentric action understanding. Comments: Accepted at the CONTEXTUS Workshop, ECCV 2026 Subjects: Computer Vision and Pattern Recognition (cs.CV) Cite as: arXiv:2608.20157 [cs.CV] (or arXiv:2608.20157v1 [cs.CV] for this version) https://doi.org/10.48550/arXiv.2608.20157 Focus to learn more arXiv-issued DOI via DataCite (pending registration) Submission history From: Marko Haralović [view email] [v1] Thu, 20 Aug 2026 15:15:08 UTC (33,790 KB)
[CV-12] Artificial Intelligence for Workflow Analysis in Colorectal Surgery: A Multicentric Cross-Procedural Development and Generalization Study
链接: https://arxiv.org/abs/2608.20154
作者: Pietro Mascagni,Julia Alekseenko,Pooja P Jain,Marta Goglia,Andrea Balla,Ludovica Baldari,Gianfranco Silecchia,Claudio Fiorillo,Vincenzo Tondolo,Salvador Morales-Conde,Luigi Boni,Sergio Alfieri,Nicolas Padoy
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Minimally invasive colorectal surgeries (MIS-CRS) are characterised by significant variability and inconsistent outcomes. ColoWorkflow, a tool for the video-based assessment (VBA) of MIS-CRS workflow, was recently validated. However, manual VBA is time-consuming, limiting implementation. This study presents AI-ColoWorkflow, a deep learning model for automated surgical workflow analysis across MIS-CRS. Operative videos of MIS-CRS were collected from 4 centres and a publicly available dataset. Phases and steps were manually annotated according to ColoWorkflow. A deep learning model combining a fine-tuned DINOv3 vision transformer for per-frame visual feature extraction with a hierarchical multi-stage temporal convolutional network was jointly optimized for phase and step recognition. The model trained on pooled multicentric data, namely AI-ColoWorkflow was compared against centre-specific and procedure-specific models on a held-out test set. The following metrics were used for evaluation: macro F1 score, balanced accuracy, precision, and recall. AI-ColoWorkflow achieved a macro F1 of 73.01% \pm 10.27 (balanced accuracy 73.43%) for phase recognition and 39.82% \pm 7.06 (balanced accuracy 38.65%) for step recognition. The global model outperformed centre- and procedure-specific models in most experiments except procedure-specific step recognition. In the generalization analysis, mean F1 was 48.42% for phase recognition. AI-ColoWorkflow can reliably recognize MIS-CRS phases. A single model trained on pooled, multicentric, multi-procedural data generalises at least as well as and often better than centre- or procedure-specific models for phase recognition in MIS-CRS, while procedure-specific step models retain advantages for certain procedure types, motivating hybrid training strategies for future surgical AI development.
[CV-13] PelviNeXt: A Modality-Agnostic Hybrid Network for Pelvic Imaging in Womens Health MICCAI
链接: https://arxiv.org/abs/2608.20144
作者: Siam Tahsin Bhuiyan,Rashedur Rahman,Sefatul Wasi,Halima Khatun,Ashraful Islam,AKM Mahbubur Rahman,Saadia Binte Alam,M Ashraful Amin
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: Accepted at MICCAI CAPI-WOMEN 2026
Abstract:Women’s health remains substantially under-resourced in medical imaging research, with pelvic pathologies such as polycystic ovary syndrome (PCOS) and pelvic fracture both suffering from a scarcity of public, well-annotated benchmark data despite their clinical importance. We introduce PelviNeXt, a modality-agnostic hybrid architecture combining a dense convolutional feature extractor, hierarchical channel-spatial attention (H-CBAM), a multi-scale fusion module (MSFM), and talking-heads multi-head self-attention (TH-MHSA), applied without modification to both pelvic ultrasound and X-ray inputs. While benchmarking PelviNeXt on PCOSGen, the only gynaecologist-annotated public PCOS ultrasound dataset, we identified extensive exact and near-duplicate contamination within and across the dataset. We audit this contamination via perceptual hashing, publicly release a deduplicated version of the dataset, and establish the first integrity-audited evaluation protocol and baseline for PCOSGen under 5-fold cross-validation. On the only publicly available pelvic fracture X-ray dataset (PXR150), PelviNeXt exceeds previously reported state-of-the-art results across accuracy, recall, specificity, and AUROC. Ablation studies confirm that each architectural component contributes to performance on both tasks. Our results demonstrate that a single architecture, applied without task-specific modification, can serve as a reliable foundation for pelvic imaging across modalities in data-scarce, under-researched areas of women’s health.
[CV-14] DPC-Net: Dual-Prior Collaborative Network for All-in-One Image Restoration
链接: https://arxiv.org/abs/2608.20141
作者: Zhaokun He,Kangbiao Shi,Axi Niu,Jian Jin,Peng Wu,Wei Dong,Qingsen Yan
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:All-in-One Image Restoration (AiOIR) aims to handle diverse degradations within a unified model. However, existing methods often overlook image semantics in degradation modeling and lack low-level visual priors during reconstruction, leading to structural distortions and semantic inconsistencies. To address these issues, we propose a novel Dual-Prior Collaborative Network (DPC-Net), which achieves high-quality restoration by jointly exploiting degradation-semantic coupled priors and low-level visual priors. Specifically, degraded images are fed into a Degradation-Aware Network (DAN) to extract degradation-semantic coupled features. To this end, a Vision-Language Model (VLM) supervises DAN by constraining its features distribution, introducing image semantics into the encoding of degradation patterns. A Degradation-Semantic Modulation Module (DSMM) further translates this guidance into degradation-semantic coupling and propagates coupled representations to the decoder. During decoding, knowledge bases provide low-level visual priors, and the Dual-Prior Collaborative Reconstruction Module (DPCR) integrates dual-prior information to guide degradation removal while preserving structure and semantics, producing high-fidelity restored images. Extensive experiments on multiple restoration benchmarks demonstrate that DPC-Net achieves superior performance against state-of-the-art AiOIR methods.
[CV-15] Feature Evolution and Migration during Vision Transformer Training CIKM2026
链接: https://arxiv.org/abs/2608.20134
作者: Joonas Järve,Halil Ibrahim Aysel,Tarun Khajuria,Meelis Kull
类目: Computer Vision and Pattern Recognition (cs.CV); Machine Learning (cs.LG)
备注: Accepted to CIKM 2026
Abstract:We present a novel view on feature evolution in Vision Transformers (ViTs) by visualizing the training process over two dimensions – network depth (layer) and training time (epochs). We employ Sparse Autoencoders (SAEs) to extract candidate sparse features from CLS-token representations and compare their activation profiles across epoch–layer pairs. This allows us to study feature-level dynamics that are not directly visible from representation-level similarity measures. Furthermore, we demonstrate how this framework of feature evolution allows us to describe feature migration, the change in the layer where a feature is most detectable during training. Our experiments show that migration is concentrated early in training, occurs more often toward earlier layers than toward deeper layers, and declines as feature organization stabilizes. We further find that deeper layers stabilize earlier and more strongly than shallow layers. The results show that our approach can be employed as a tool for understanding how ViTs learn and evolve.
[CV-16] ID-VTG: Image-Disambiguated Video Temporal Grounding
链接: https://arxiv.org/abs/2608.20127
作者: Minghang Zheng,Jingli Wei,Hongyi Yang,Yang Liu
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: ACM-MM 2026
Abstract:Video Temporal Grounding (VTG) faces significant challenges when natural language queries must distinguish between multiple events involving visually similar entities, particularly when relying on fine-grained visual attributes that are difficult to describe accurately in words alone. To address this, we introduce Image-Disambiguated Video Temporal Grounding (ID-VTG), a task that leverages multimodal queries combining a reference image and a text description to precisely localize segments where a specific instance performs a described action. To facilitate research, we construct two benchmarks: IDVTG-Gym, focusing on fine-grained, compositionally ordered gymnastics actions with athletes in similar uniforms; and IDVTG-InternVid, an open-world dataset featuring diverse entities (e.g., humans, animals, fictional characters) and significant temporal distractors. Methodologically, we propose the Visually-Guided Disambiguation Aggregation (VGD-Agg) framework based on a dual-branch fast-slow architecture. The fast branch efficiently generates preliminary event proposals, while the slow branch performs fine-grained frame-level matching between video frames and the reference image. We enhance discriminability via two learnable tokens: a Compare Token, which represents hard negatives to probe for the presence of the target instance (as referred to by the query image), and a Depress Value, which represents text-irrelevant events. Proposals that the Compare Token identifies as lacking the target instance are pushed toward the Depress Value, thus easing disambiguation via the text query. Extensive experiments validate our approach, which achieves state-of-the-art results on the proposed benchmarks. Code is available at this https URL.
[CV-17] ArmorOCR: Grounded Adversarial Visual Perception via Observation-Transferred Self-Distillation
链接: https://arxiv.org/abs/2608.20122
作者: Linhan Cao,Siyuan Li,Jun Lan,Liangbo He,Guannan Li,Xiaolei Huang,Jun Jia,Shuheng Zhou,Huijia Zhu,Weiqiang Wang,Wei Sun
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Large multimodal models (LMMs) have demonstrated strong OCR recognition capabilities, yet remain vulnerable to adversarial visual text that is readable to humans but challenging for models to localize and recognize. Existing OCR benchmarks mainly focus on natural or document-style text, while adversarial OCR evaluations remain limited in scale, task coverage, or region-aware evaluation. In this paper, we formulate adversarial OCR as a \textbfgrounded OCR perception task and introduce \textbfAdvSpot, the first benchmark for grounded adversarial OCR evaluation. AdvSpot comprises 390 images with region-level annotations, spanning 5 primary categories and 13 fine-grained adversarial OCR types. To address this challenge, we propose \textbfArmorOCR, a two-stage training framework for robust adversarial OCR perception. ArmorOCR first acquires missing adversarial OCR perception from privileged transformed observations through On-Policy Self-Distillation (OPSD), and then refines grounded OCR perception through Group Relative Policy Optimization (GRPO) with task-conditioned rewards for localization, recognition, full spotting, and visual question answering (VQA). Experiments on our AdvSpot, other adversarial OCR benchmarks, and general OCR benchmarks demonstrate that ArmorOCR consistently improves adversarial OCR perception while preserving competitive general OCR capability.
[CV-18] BeyondMasks: Evaluating Causal and Physical Consistency in Video Object Removal ECCV2026
链接: https://arxiv.org/abs/2608.20107
作者: Yigit Ekin,Enes Sanli,Aykut Erdem,Erkut Erdem,Aysegul Dundar
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: ECCV 2026 Project Page: this https URL
Abstract:Recent advances in generative video models have significantly improved visual realism in video object removal, yet evaluation protocols still focus on masked region fidelity, treating removal as local inpainting. In real scenes, object removal is a causal intervention: eliminating an object also requires removing its induced physical effects, such as shadows, reflections, illumination changes, translucency, and dynamic traces. Existing benchmarks lack aligned clean references or remain limited to simplified synthetic settings, preventing systematic evaluation of causal consistency. We introduce BeyondMasks, a paired benchmark for causally consistent video object removal, consisting of temporally aligned synthetic and real world video pairs with clean background references. The dataset spans diverse photometric, geometric, volumetric, and dynamic interactions, and supports both mask based and instruction driven editing. We further propose CORE, a structured vision language model based evaluation protocol that jointly measures object disappearance and after effect consistency, aligning more closely with human judgments than existing metrics. Benchmarking state of the art methods reveals systematic failures in removing secondary physical effects despite high masked region fidelity, exposing a gap between visual plausibility and causal correctness. BeyondMasks reframes video object removal as causal scene consistency rather than local reconstruction and provides a unified framework for its evaluation.
[CV-19] Structured Affinity for Unsupervised Visual Class-Incremental Memory in Deep Artificial Immune Networks
链接: https://arxiv.org/abs/2608.20104
作者: Siphesihle Sithungu
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
备注: 18 pages, 3 figures
Abstract:Artificial immune networks (AINs) are naturally memory-forming systems, but conventional visual AINs often rely on flattened vector affinity that ignores spatial structure. This paper studies whether structured, gradient-free immune affinity can make Deep AINs viable as replay-free visual class-incremental representation-memory learners. Visual B-cells are formalized as structured templates, including shifted-template affinity, zero-normalized cross-correlation (ZNCC) filters, and feature-map binding profiles. A repertoire is treated both as memory and as a representation-inducing basis, while depth is obtained by passing binding-profile response maps to subsequent immune layers. The resulting Deep AIN exhibits adaptive latent coordinate reorganization: as new classes arrive, the binding-profile space evolves while retaining recoverable structure for earlier classes. Experiments on sklearn digits, MNIST, Fashion-MNIST, and KMNIST show that preserving response maps is critical. Scalar binding-profile variants underperform, whereas feature-map Deep AINs learn class-discriminative visual memory without replay, label-driven immune updates, or backpropagation through the immune layers. On sklearn digits, downstream probes fitted on the learned binding profiles reach 0.939 final balanced accuracy with logistic regression and 0.902 with 1-nearest-neighbour after all ten classes are encountered, with initial-class retention of 0.978. Adaptive layer-wise scale calibration further improves the two-layer feature-map Deep AIN to 0.978 balanced accuracy. With the same calibration rule, Fashion-MNIST reaches 0.814 and KMNIST reaches 0.853. These probes are external validation tools, not components of the AIN. The results identify structured affinity, response-map preservation, adaptive latent reorganization, and layer-wise scale calibration as key mechanisms for replay-free visual immune memory.
[CV-20] HandMvNet: Real-Time 3D Hand Pose Estimation Using Multi-View Cross-Attention Fusion
链接: https://arxiv.org/abs/2608.20093
作者: Muhammad Asad Ali,Nadia Robertini,Didier Stricker
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: 8 pages, 7 figures. Published at VISAPP 2025
Abstract:In this work, we present HandMvNet, one of the first real-time method designed to estimate 3D hand motion and shape from multi-view camera images. Unlike previous monocular approaches, which suffer from scale-depth ambiguities, our method ensures consistent and accurate absolute hand poses and shapes. This is achieved through a multi-view attention-fusion mechanism that effectively integrates features from multiple viewpoints. In contrast to previous multi-view methods, our approach eliminates the need for camera parameters as input to learn 3D geometry. HandMvNet also achieves a substantial reduction in inference time while delivering competitive results compared to the state-of-the-art methods, making it suitable for real-time applications. Evaluated on publicly available datasets, HandMvNet qualitatively and quantitatively outperforms previous methods under identical settings. Code is available at this http URL.
[CV-21] V-REX: Efficient Specialist VLM Training for Veterinary X-Rays
链接: https://arxiv.org/abs/2608.20069
作者: Tim Elsner,Nicole McNally,Andre Dourson,Michael Fitzke
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:While generalist VLMs are expensive to train, creating domain experts is widely assumed to require fine-tuning increasingly large foundation models. We show that, in veterinary radiology, this assumption is misguided. By rethinking the entire VLM pipeline - from text tokenisation and pre-training to grounding and inference - we demonstrate that careful engineering can yield models that outperform much larger foundation models from scratch, without relying on any other data. Our approach introduces new strategies for generative pre-training and grounding that improve training efficiency, increasing data utilisation and downstream performance. Using only a fraction of the parameters, data, and compute of contemporary generalist models, we develop the first VLM capable of generating diagnostic reports for veterinary radiographs, surpassing open foundation models on this task by significant margin.
[CV-22] Gravity-aware partially calibrated absolute pose estimation from affine- or rotation-covariant features
链接: https://arxiv.org/abs/2608.20056
作者: Marcus Valtonen Örnhag,Alberto Jaenal,Stefan Adalbjörnsson
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: European Conference on Computer Vision 2026
Abstract:Inertial measurement units (IMUs) are now standard in most consumer devices, such as smartphones, drones, and extended reality (XR) headsets. By fusing visual and inertial data, localization systems gain significantly in speed and robustness compared to vision-only or IMU-only approaches. However, traditional pose estimation methods fail to utilize the local geometric information embedded in feature descriptors like SIFT. Recent work has proved the advantages of leveraging this information for relative and absolute pose estimation, but its application to partially calibrated absolute pose estimation remains unexplored. In this paper, we derive novel constraints for joint estimation of absolute pose and focal length, making use of a gravity vector obtained from IMU data and the feature-induced local geometry, which we use to construct two efficient solvers: UP1PfAC, that operates given a single affine correspondence and UP2PfORI, which requires two orientation-covariant features. Unlike traditional, semi-calibrated absolute pose methods requiring four point correspondences, our solvers benefit from fewer samples and lower computational cost, simplifying robust estimation in modern RANSAC-like frameworks. We evaluate the proposed solvers against the state-of-the-art on large-scale public datasets and demonstrate that our method achieves fast and accurate localization and focal length estimation.
[CV-23] An Inclusive and Lightweight Approach to Federated Continual Learning for Cultural Heritage
链接: https://arxiv.org/abs/2608.20038
作者: Ioannis Theologitis,Debin Meng,Stylianos Eleftheriadis,Vasileios Lolis,Konstantinos Votis
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Computer Vision and Pattern Recognition (cs.CV)
备注: 7 pages, 3 figures, Accepted at the 2026 IEEE International Conference on Cyber Humanities (IEEE-CH 2026), Venice, Italy, September 7–9, 2026. Accepted author manuscript. Copyright 2026 IEEE
Abstract:Artificial intelligence can support cultural heritage and digital humanities through large-scale retrieval and analysis of digitized collections. However, cultural heritage data are often distributed across institutions, constrained by ownership and access restrictions, and continuously evolving over time. Federated Continual Learning (FCL) is well suited to this setting, as it enables models to learn from distributed and sequential data without sharing raw collections. In this paper, we propose FedCurv-DR, a lightweight, regularisation-based FCL strategy. The method accumulates parameter-importance estimates across clients and experiences to protect learned knowledge, while updating them only at fixed intervals to minimize communication and computation overhead. We evaluate FedCurv-DR in a continual learning scenario using the WikiArt image dataset for genre classification with evolving styles, reporting performance, energy, and fairness metrics. Our results show that FedCurv- DR reduces forgetting and balances performance, fairness, and energy efficiency for sustainable AI in cultural heritage.
[CV-24] From Street View Imagery to Street Quality Indicators: Vision Language Inference for the Suburban 15-minute City
链接: https://arxiv.org/abs/2608.20026
作者: Joan Perez,Giovanni Fusco
类目: Computer Vision and Pattern Recognition (cs.CV); Machine Learning (cs.LG)
备注: 16 pages, 4 figures
Abstract:Streetscape quality has become a central concern in contemporary urban planning, particularly within the framework of the pedestrian-friendly 15-minute city, where walkability and public-space quality are increasingly recognized as key determinants of urban performance. However, assessing streetscape qualities across large suburban and peri-urban territories remains challenging due to the time and resource demands of conventional field surveys. This paper presents a planning-oriented assessment of streetscape qualities in the north-eastern periphery of Nice (France) using the latest release of SAGAI (Streetscape Analysis with Generative AI), an open-source workflow that leverages vision-language models (VLMs) for large-scale streetscape analysis from Google Street View imagery. The new release addresses limitations of the original framework through improved image acquisition, geographically consistent view generation, support for multiple VLM architectures, consensus-based inference, and an integrated analytical environment. The workflow is applied to several thousand street-level observations to evaluate qualities relevant to pedestrian-friendly urban environments: sidewalk presence, pedestrian entrance density, and vegetation. The resulting maps reveal that the desired streetscape qualities characterize only a fraction of today’s suburban streetscapes, mainly in compact developments and traditional suburban faubourgs, while they are particularly lacking on residential hills. The analysis demonstrates the potential of contemporary VLMs to support urban diagnostics in extensive suburban territories where fieldwork would be prohibitively time-consuming. Beyond the case study, the paper illustrates how recent advances in vision-language models can contribute to evidence-based planning by enabling scalable, flexible, and interpretable assessments of urban public-space quality.
[CV-25] Manifold Drift in Flow Preference Optimization: A Root Cause of Reward Hacking
链接: https://arxiv.org/abs/2608.20011
作者: Yansen Han,Shengyi Liao,Yuanxing Zhang,Pengfei Wan,Tao Lin
类目: Artificial Intelligence (cs.AI); Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Preference optimization is a standard alignment method for generative models, yet extending it to continuous-time dynamics remains non-trivial. In flow matching, reward-driven updates modify transport trajectories without an inherent constraint to the pretrained data manifold and can move terminal samples off the pretrained support. We formalize this failure mode as manifold drift. Theoretically, we show that optimal flow matching recovers the terminal data distribution, whereas a preference update leaves the pretrained manifold whenever its induced terminal displacement has a nonzero normal component. As a remedy, we propose ThermoDPO, a temperature-controlled objective that anchors pairwise preference optimization on preferred samples. Across temperature regimes, this objective connects rejection sampling fine-tuning and FlowDPO and controls a pointwise reconstruction-based surrogate for manifold distance. To counteract diminished signals at low temperatures, we further introduce a weighted variant, ThermoDPO-weighted. On the main toy benchmark, ThermoDPO-weighted attains a StrictScore of 0.899, compared with 0.629 for FlowDPO and 0.857 for FlowDPO+RFT. On SD3.5-M at CFG = 4.5, it improves OCR by 47.5% and the average of four metrics by 16.0%.
[CV-26] Point-Based 3D Reconstruction from Sparse Views under Known Illumination
链接: https://arxiv.org/abs/2608.20000
作者: Magnus Kaufmann Gjerde,Joakim Bruslund Haurum,Jeppe Revall Frisvad,Markus Worchel,J. Andreas Bærentzen,Thomas B. Moeslund
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Sparse view 3D reconstruction is commonly addressed with neural implicit surfaces or dense point-based representations such as Gaussian splatting. Surface-aware splatting methods improve extracted geometry through oriented primitives and regularization, while RadiosityGS incorporates differentiable light transport through a radiosity inspired finite-element surfel formulation. We propose a differentiable point rendering method based on opacity-bearing beta surfels. An opacity explicit adjoint light transport formulation provides gradients for surfel geometry and appearance parameters, allowing physically based light transport to constrain reconstruction. Across five synthetic objects reconstructed from ten posed views, our method achieves the lowest mean symmetric Chamfer distance among the evaluated baselines and reduces mean Chamfer distance by 28.5% relative to the strongest point-based baseline while using only 267 surfels on average, approximately ~161 fewer primitives. Directional Chamfer results further show improved accuracy and competitive completion relative to related point-based methods. These results show that, in the controlled direct illumination setting, compact beta surfels combined with transport-based optimization can recover surfaces without relying on the tens to hundreds of thousands of primitives used by the evaluated baselines. Subjects: Computer Vision and Pattern Recognition (cs.CV) Cite as: arXiv:2608.20000 [cs.CV] (or arXiv:2608.20000v1 [cs.CV] for this version) https://doi.org/10.48550/arXiv.2608.20000 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
[CV-27] STEP: Score-Based Temporal Energy for Human Pose Video Anomaly Detection ECCV2026
链接: https://arxiv.org/abs/2608.19987
作者: Jakub Micorek,Mateusz Koziński,Horst Possegger
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: Accepted to ECCV 2026. Project page: this https URL | Code: this https URL
Abstract:Skeleton-based Video Anomaly Detection (VAD) offers a robust, privacy-preserving solution for identifying abnormal behaviors. To model the distribution of normal static and moving poses, recent methods train Energy-Based Models (EBMs) via Denoising Score Matching (DSM). However, directly injecting noise, required for training, into raw joint coordinates creates physically impossible poses, and this structural collapse severely worsens as the temporal window expands. To address this, we introduce STEP, a simple framework that utilizes Principal Component Analysis (PCA) to project pose sequences into a compact, whitened PC-space. Learning the data density within this well-behaved PC-space ensures that the injected noise translates into physically plausible variations, which allows the model to process longer video sequences without the performance collapse of raw coordinate baselines. Additionally, to mitigate inherent pose estimation inaccuracies arising from occlusions or motion blur, we integrate a sequence-level weighting mechanism based on the estimator’s confidence scores. Operating at real-time computational efficiency, our simple and lightweight framework outperforms the previous skeleton-based state-of-the-art by 12.2% (90.1% AUROC) on the challenging UBnormal dataset and achieves highly competitive results by improving on the ShanghaiTech benchmark.
[CV-28] Open-Vocabulary 3D Object Detection with Co-Distillation Discovery and Dual Guidance Robust Training ECCV26
链接: https://arxiv.org/abs/2608.19973
作者: Shangbo Yuan,Jie Xu,Xiaofeng Zhu,Na Zhao
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
备注: Accepted by ECCV26
Abstract:Recently, open-vocabulary 3D object detection (3D-OVD) has gained increasing attention for its ability to detect unseen objects in 3D scenes. Existing approaches typically adopt a two-stage pipeline that first discovers novel objects using foundation models and then trains a 3D-OVD model based on these discovered objects. Although effective, this pipeline often suffers from inaccurate localization and mismatched classification during the discovery stage, which subsequently limits the performance of the model training stage. To address these limitations, we advocate for improving both the reliability of novel object discovery and the robustness of model training, and propose an innovative framework. Specifically, for reliable discovery, our co-distillation strategy distills high-quality novel objects by applying Hungarian matching over a comprehensive score that incorporates geometric consistency, structural objectness, and semantic certainty. To enhance robust model training, we further propose a dual-guidance learning scheme, incorporating a scene-awareness-guided uncertainty regularization for the regression head and an LLM-guided hierarchical alignment for the classification head, effectively mitigating the negative effects of imprecise 3D bounding boxes and semantic ambiguity. Extensive experiments on SUN RGB-D and ScanNetV2 demonstrate that our method achieves significant performance gains over state-of-the-art approaches. Code is available at this https URL
[CV-29] PVRA: A Pointwise Key-point Voting Framework for Robotic Assembly
链接: https://arxiv.org/abs/2608.19968
作者: Kulunu Samarawickrama,Roel Pieters
类目: Robotics (cs.RO); Computer Vision and Pattern Recognition (cs.CV)
备注: 14 pages, 3 figures. Accepted for presentation at the European Conference on Robotics (ECoR) 2026
Abstract:Modern computer vision has enabled partial autonomy in robotic assembly manipulation. However, performing autonomous manipulation of a progressive assembly demands a more specific set of skills, in addition to perceiving the objects. Through a comparative analysis of research in the associated domains, we deduce that object-centric perception must advance towards learning assembly dependencies to predict meaningful actionable outputs for autonomous assembly manipulation. Subsequently, we present a 3D keypoint-based modular learning framework to learn assembly dependencies to infer actionable outputs given a RGB-D input of an assembly scene. We train and evaluate our trained network on an assembly pose estimation dataset and compare it against object-centric baselines with an augmented set of metrics for progressive assemblies.
[CV-30] Flow Matching Meets 3D Curvilinear Structure Segmentation in Medical Imaging MICCAI
链接: https://arxiv.org/abs/2608.19965
作者: Sidi Mohamed Sid’El Moctar,Nicolas Vitry,Hélène Bouvrais
类目: Computer Vision and Pattern Recognition (cs.CV); Machine Learning (cs.LG); Quantitative Methods (q-bio.QM)
备注: International Workshop on Machine Learning in Medical Imaging (MLMI 2026) @ MICCAI
Abstract:Segmentation of curvilinear anatomical structures in 3D medical images remains challenging due to complex topology, severe class imbalance, weak contrast, and large variations in structure morphology. While deep learning approaches for 3D curvilinear segmentation have been proposed, they are often tailored to specific anatomies or modalities, limiting generalization across clinical settings and leaving room for improvement. Recent generative models have shown the benefits of iterative prediction for structured segmentation tasks, yet diffusion-based methods suffer from computationally expensive sampling, hindering their use on high-resolution 3D volumes. We present 3D-CurvSegFlow, a flow matching-based model for 3D curvilinear structure segmentation. The model learns a continuous transformation from a simple source distribution to the target vascular representation, enabling progressive refinement of complex curvilinear geometries with efficient inference. We evaluate our method on Three public challenging datasets covering distinct anatomies and modalities: portal vein, cerebral vessel, and coronary arteries. Using a common architecture and training strategy across all tasks, our method outperforms general-purpose and vessel-specific approaches, with strong preservation of thin branches and vascular continuity. This work not only advances the state-of-the-art in 3D curvilinear segmentation but also opens new avenues for efficient, generalizable, and clinically applicable methods in medical image analysis.
[CV-31] AvatarDynamizer: From Static to Dynamic Human Avatars via Generative Dynamic Textures
链接: https://arxiv.org/abs/2608.19900
作者: Guoxing Sun,Heming Zhu,Linjie Lyu,Pascal Fua,Christian Theobalt,Marc Habermann
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: Project page: this https URL
Abstract:For full-body avatars, modeling surface dynamics is crucial for overcoming the uncanny valley and achieving perceptual realism. Person-agnostic methods recover static 3D avatars from monocular images, videos, or text prompts, but their skeleton-driven animations lack realistic surface dynamics such as clothing wrinkles. In contrast, person-specific methods achieve high-quality rendering and realistic dynamics, but require expensive multi-view captures for each individual. Recent generalizable dynamic avatar methods struggle to embed surface dynamics, leading to either limited multi-view consistency or dynamic expressiveness. To this end, we propose AvatarDynamizer, a generative method that transforms an off-the-shelf static 3D avatar into a controllable, realistic, and multi-view-consistent 4D avatar. We introduce a novel texture-space surface-dynamics embedding and formulate avatar dynamics modeling as conditional texture generation. Our encoder–decoder representation embeds pose-dependent dynamics into dynamic texture maps, enabling compatibility with pre-trained video diffusion models while decoding them into 3D Gaussians for multi-view consistent rendering. Since existing datasets are limited in scale, sequence length, or motion diversity, we collect a large-scale multi-view dataset with long sequences covering diverse skeletal motions and surface dynamics. Experiments show that our method effectively animates static avatars with faithful surface dynamics and outperforms competing generalizable methods in visual fidelity, especially under limited dynamic training data.
[CV-32] Unified and Efficient Point-Line Local Features
链接: https://arxiv.org/abs/2608.19894
作者: François Costa,Raphael Kreft,Eckhard Goedeke,Felix Möller,Hardik Shah,Ramanathan Rajaraman,Shaohui Liu,Rémi Pautrat,Marc Pollefeys
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Multi-view computer vision pipelines typically rely on accurate sparse keypoints and robust descriptors. While incorporating line features has shown clear benefits for matching and pose estimation, existing point-line approaches remain inefficient: they detect points and lines separately, use increasingly heavy networks, and depend on CPU-bound heuristics that hinder real-time performance. We introduce a Unified Efficient Points and Lines (UPAL) feature extractor that jointly extracts keypoints, line segments, and feature descriptors within a single lightweight architecture. A shared backbone provides common representations that feed different branches for point and line features. Line segments are recovered through an accelerated post-processing stage, an enhanced and highly efficient variant of the LSD algorithm. UPAL matches or exceeds state-ofthe-art performance in both point and line applications while significantly reducing computational cost, achieving, for instance, a 4x speedup and 10x smaller memory footprint over the ALIKED + DeepLSD pipeline. Code is publicly available at this https URL.
[CV-33] DIFFCZSL: Compositional Zero-Shot Learning Regularized by Diffusion Representations
链接: https://arxiv.org/abs/2608.19871
作者: Hangyu Tian,Zhenqi He,Yanghao Wang,Long Chen
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Compositional Zero-Shot Learning (CZSL) aims to recognize unseen attribute-object compositions by leveraging knowledge of primitive concepts learned from seen compositions. Although recent works achieve impressive performance in CZSL by leveraging large vision-language models, they primarily rely on discriminative representations that may not explicitly preserve the structured relationships between primitive concepts and their compositions. Motivated by the recent success of diffusion-based classifiers and their competitive performance relative to discriminative models, we investigate whether intermediate diffusion representations can provide complementary cues for CZSL. To this end, we propose DIFFCZSL, a diffusion-augmented framework that injects generative priors from pre-trained diffusion models into CLIP-based CZSL pipelines. We extract intermediate diffusion representations and project them into the CLIP embedding space to provide auxiliary supervision on both image and text modalities. Through contrastive alignment between CLIP embeddings and diffusion features during training, our method encourages the embedding geometry toward richer composition-aware semantics, while introducing no additional cost at inference time. Extensive experiments on three public CZSL benchmarks demonstrate consistent improvements over strong CLIP-based baselines under both closed-world and open-world settings. Our results highlight the complementary strengths of generative diffusion representations and discriminative vision-language models for compositional generalization.
[CV-34] A 360-Degree Vision Dataset for Learning Yaw Control on GPS-Denied Micro-UAVs in Disaster-Response-Relevant Environments
链接: https://arxiv.org/abs/2608.19866
作者: Niklas Voigt,Hartmut Surmann
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: Accepted at the 2026 IEEE/ASME International Conference on Advanced Intelligent Mechatronics (AIM), Genoa, Italy, July 6-10, 2026
Abstract:This paper presents a novel data-driven approach to camera-based autonomy for micro-drones in GPS-denied, radio-challenging indoor environments. The target application is disaster and emergency response, where micro-UAVs can provide rapid situational awareness in hazardous settings such as firefighting and chemical, biological, radiological, and nuclear (CBRN) incidents while reducing risk for human responders. When the communication link is lost, the micro-drone uses a learned yaw controller to autonomously navigate toward open space, preserving onboard sensor data that would otherwise be lost with the vehicle. A custom micro-drone equipped with a 360-degree camera was used to record diverse industrial, underground, and training scenarios representative of communication-denied field operations. We introduce a preprocessing pipeline that converts equirectangular 360-degree footage into planar front views and dynamically generates image-label pairs for AI training. We then train and compare multiple convolutional neural network variants that predict a continuous yaw command from a single monocular view. Evaluation on a held-out test set confirms the feasibility of the learned yaw-prediction approach. A semi-autonomous real-world test further demonstrates the practicality of the method while revealing key failure modes, particularly reflections and glare.
[CV-35] AutoLumNet: Monotone Optimal Transport for Single-Shot Exposure Correction
链接: https://arxiv.org/abs/2608.19860
作者: Airin Akter Tania,Md Raihan Khan,Mohiuddin Ahmad
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Single-shot exposure correction aims to map an arbitrarily degraded image—whether under-exposed, over-exposed, or a spatial mixture of both—to a well-exposed output from a single capture. We present AutoLumNet, a framework that decomposes this task into a global monotone tone curve and a bounded local residual, making the global component the locus of formal guarantees. The tone curve is parameterized as the normalized cumulative integral of a strictly positive density, ensuring strict monotonicity by construction rather than by penalty. We prove that this parameterization (i)~preserves the pairwise luminance ordering of all pixels and all spatial extrema unconditionally, and (ii)~is dense in the space of valid tone corrections, containing the one-dimensional optimal-transport map from the input to any target luminance distribution. A differentiable sorted-sample Wasserstein-2 objective drives the learned curve toward the OT optimum during training. Spatially varying effects that the global map provably cannot address—local shading, chrominance shifts, and clipped-region restoration—are handled by a bounded residual decoder with dual-branch convex fusion, for which we provide an explicit sufficient condition for local order preservation. Experiments on five benchmarks (MSEC, SICE, LCDP, LOL-v1, LOL-v2-real) show that AutoLumNet achieves state-of-the-art PSNR and SSIM across both under- and over-exposure regimes at 11.2,ms per frame, and generalizes zero-shot to pure low-light benchmarks without retraining. To our knowledge, AutoLumNet is the first exposure-correction method to unite structural monotonicity, optimal-transport optimality, and bounded local adaptivity within a single trainable architecture. Code is available at this https URL.
[CV-36] Core-KAN: Continuous Vision Kernels with Kolmogorov-Arnold Networks
链接: https://arxiv.org/abs/2608.19817
作者: Lan Guo,Mengling Li,Haoran Li,Jun Shen,Yuanbo Jiang,Qingguo Zhou,Binbin Yong
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
备注:
Abstract:Conventional convolutional kernels are typically defined on fixed discrete grids, limiting their ability to accommodate heterogeneous local structures. Existing adaptive operators improve flexibility but often couple geometric scale variation with content-dependent filtering, while incurring high computational cost from per-location kernel generation. To decouple geometric scale adaptation from content-dependent filtering while avoiding expensive per-location kernel generation, we propose Continuous Relative-scale KAN (Core-KAN), a relative-scale-conditioned continuous convolution operator. Core-KAN maps input features into a compact latent basis space and uses a lightweight scale controller to predict local scales relative to an exponential moving average reference. A KAN-based generator represents depth-wise kernel bases as continuous coordinate functions, allowing the operator to synthesize spatial filters at arbitrary resolutions rather than being confined to a fixed lattice. Instead of synthesizing independent kernels at every location, it constructs a compact bank of scale-conditioned kernel responses and interpolates them according to the predicted local scale map. An independent mixing controller further combines the interpolated basis responses based on local content, explicitly decoupling geometric scale adaptation from content-dependent filtering. Together with lightweight pointwise projections, this design forms a low-rank dynamic convolution that scales efficiently with kernel size and integrates readily into hierarchical vision backbones. Experiments across three representative vision tasks show Core-KAN consistently outperforms strong convolutional and dynamic-kernel baselines with only marginal parameter and computational overhead, offering an efficient, general framework for continuous, scale-adaptive convolution.
[CV-37] When Saying No Makes Better Videos: Designing Dual Gatekeeping for Pedagogically Grounded AI Content Creation
链接: https://arxiv.org/abs/2608.19812
作者: Yearim Kim,Njun Baek,Nojun Kwak
类目: Artificial Intelligence (cs.AI); Computer Vision and Pattern Recognition (cs.CV)
备注: 4 pages, 1 figure. Presented at the CHI 2026 Workshop on Understanding and Engaging Critical Resistance to AI in Education
Abstract:To prevent the adoption of aesthetically polished but pedagogically flawed AI content, we study a video authoring pipeline featuring two layers of structured refusal. The first layer empowers educators to iteratively reshape AI scripts based on multimedia learning theory, while the second employs automated metrics to flag violations in instructional coherence and narrative-visual synchronization. While neither layer is exhaustive, their synergy ensures that principled resistance–the act of deferring AI output until it meets rigorous standards–becomes a catalyst for higher quality. Evaluation combining a study with 23 educators across 3 topics and automated metrics across 7 topics drawn from established science and philosophy curricula shows that both layers independently improve the same instructional dimensions, suggesting that thoughtful resistance and generative AI are not opposites but partners.
[CV-38] Coupled Optimal Transport with Landmark Constraints
链接: https://arxiv.org/abs/2608.19783
作者: Xiang Gu,Jian Sun,Zongben Xu
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Existing optimal transport (OT) models primarily seek an OT map or plan between distributions by minimizing a prescribed transport cost or distortion. However, minimizing transport cost or distortion alone may fail to identify a geometrically meaningful transformation between the two distributions. To address this limitation, this paper proposes a novel coupled OT framework that leverages a small number of annotated landmarks to guide the recovery of an underlying deformation governing the distribution transformation. The coupled OT framework integrates the optimization of the transport plan and the deformation field into a unified model, where the landmark-guided deformation field and the cost-driven transport plan are coupled through a mutual-consistency constraint. As a result, the deformation is jointly determined by the annotated landmarks and cost-driven distribution matching. The proposed framework provides a principled connection between landmark-based registration and transport-based distribution matching, enabling the recovery of transport maps from sparse geometric supervision. We establish the well-definedness of the proposed model in a general variational setting and develop a finite-element-based numerical algorithm for computation whose convergence properties are systematically analyzed. The practical effectiveness of the proposed approach is verified in shape matching.
[CV-39] Far from the Crowd: Scalable Self-Supervised Learning via Geographic Isolation ECCV2026
链接: https://arxiv.org/abs/2608.19766
作者: Daniele Rege Cambrin,Francesco Rossi,Mattia Varile
类目: Computer Vision and Pattern Recognition (cs.CV); Machine Learning (cs.LG)
备注: Accepted to ECCV 2026 TerraBytes Workshop
Abstract:Self-supervised pretraining on remote sensing imagery typically treats all samples as equally informative, despite large variability in geographic and visual structure. We propose a curriculum learning strategy for self-supervised Earth observation that ranks samples by geographic isolation, a label-free proxy derived entirely from geolocation metadata already present in geospatial datasets, requiring no image decoding, no model feedback, and no manual annotation. Unlike visual complexity proxies, it scales as O(D log D) with dataset size D and is well-defined for both contrastive and reconstructive objectives. We integrate the proposed measure into MoCoV2 and MAE pretraining and evaluate across three downstream tasks from CopernicusBench (BigEarthNet, DFC-2020, LCZ). Our curriculum reaches baseline final-epoch performance using as few as 20% of the training budget (MAE) and at most 40% (MoCo) of the training budget, and improves final downstream performance by up to +5 mAP on BigEarthNet, with gains of 1-5 points across benchmarks, matching visual-complexity curricula while reducing pre-computation cost by more than 140x (4 s vs. 568 s on SSL4EO). A CKA and effective-rank analysis further reveals that curriculum-trained encoders develop higher-dimensional, more uniformly utilized embedding spaces throughout training.
[CV-40] Gallileo-4D: Frozen Backbone Ensemble for Dynamic 4D Reconstruction ECCV2026
链接: https://arxiv.org/abs/2608.19743
作者: Nicolò Savioli
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: Technical report for the PhysAI Dynamic 4D Reconstruction Challenge at the ECCV 2026 Workshop on Physical AI. Third of 27 teams. 14 pages, 10 figures. Code: this https URL Weights: this https URL
Abstract:We describe our entry to the PhysAI Dynamic 4D Reconstruction Challenge, which placed third of 27 teams at 0.58356 APD on the final leaderboard, without a single gradient update. This was not the plan: of thirteen fine-tuning configurations of a pre-trained 4D backbone, twelve degraded the challenge score, and eleven of those twelve improved local validation at the same time. We trace this inversion to the structure of the benchmark: only 25% of the evaluation set belongs to the data variant released for training, so updates that fit the available data damage the pre-trained features the remaining 75% relies on. Our system therefore freezes the backbone and spends its budget at inference time, fusing three decoding configurations – temporal stride-3, horizontal-flip test-time augmentation, and dense stride-1 – under a convex weighting. The ensemble recovers +0.041 APD over the frozen baseline, more than any training run achieved, at zero training cost.
[CV-41] Question-Guided Evidence Acquisition for Multimodal Visual Question Answering
链接: https://arxiv.org/abs/2608.19739
作者: Alin-Ionut Popa
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
备注:
Abstract:Multimodal LLMs can see a document, but they often can’t read it reliably. Small text, tables, visual cues, and topological elements still trip them up under direct visual inference, even when the page is already sitting in the model’s context. Most document-VQA systems treat perception as fixed: they encode the page once, ask the question, and answer from whatever the model happened to extract in that single fast pass. We think document VQA needs slower, more deliberate perception: rather than answering from one fixed encoding, the model should spend a bit of extra compute at inference time working out what to look at next, and only then answer. We build this into \textbfQ-Guide, a small agent that reads a question, works out what evidence it is still missing, and calls targeted tool(s) to recover it—reading text where text is needed, zooming in where detail is needed, or grounding a region where position matters. On DocVQA2026 and Manga109, Q-Guide outperforms both direct prompting and recent multi-agent document systems ( 65.0% vs.\ 40.0% on DocVQA2026, 32.4% vs.\ 24.4% on Manga109), and the improvement holds across three Claude backbones (Opus 4.6, Sonnet 4.6, and Opus 4.5). We find that accuracy scales with the perception budget—most of the gain appears within two to three deliberate rounds—and that the gain comes from directing perception to the right place, not from complex control logic: adding planners, routers, or multiple collaborating agents does not help.
[CV-42] Learning to Beat: Phenotype-Guided Latent Flow with Regional Motion Priors for Biventricular Motion Synthesis
链接: https://arxiv.org/abs/2608.19738
作者: Xuan Yang,Xiaohan Yuan,Hao Li,Lingyu Chen,Yanan Liu,Qingya Li,Lei Li
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
备注: 14pages, 10 figures
Abstract:Full-cycle biventricular geometry is essential for characterizing cardiac function. However, dense and temporally consistent 3D+t biventricular meshes are not routinely available, whereas end-diastolic (ED) anatomy can often be obtained reliably. We therefore investigate full-cycle biventricular motion synthesis from a single ED mesh. This task is challenging because cardiac deformation is spatially heterogeneous and phenotype dependent, while conventional global generative models often obscure localized motion patterns. In this study, we propose a region-specific and phenotype-adaptive framework that integrates motion-informed functional parcellation with conditional latent flow. A functional partition learned from reconstructed motion organizes the ventricular surface into regions with coherent dynamics and enables topology-aware regional feature exchange. A phenotype-conditioned rectified-flow model subsequently maps the ED anatomy to full-cycle motion latents through fine-grained conditioning and prototype-routed motion adapters. An optional control branch further incorporates available motion descriptors for controllable synthesis. Experiments on ACDC, M\Ms, and M\Ms-2 demonstrate consistent improvements in geometric accuracy and functional fidelity. Under ED-only synthesis, our method achieves biventricular ASSD, HD95, and vRMSE of (1.49\pm0.34)~mm, (3.77\pm1.06)~mm, and (3.31\pm1.03)~mm, respectively, outperforming all competing methods. Complementary functional and robustness evaluations further demonstrate that the synthesized sequences preserve physiologically plausible ventricular dynamics and generalize across cohorts and disease phenotypes. The code will be released publicly upon acceptance of the manuscript for publication.
[CV-43] SafeBranch: Branch-Pair Safety Alignment for Embodied Agents
链接: https://arxiv.org/abs/2608.19729
作者: Hyunse Lee,Jiwoo Jeong,Haneul Lee,Kyochul Jang,Youngjae Yu,Woojin Lee
类目: Artificial Intelligence (cs.AI); Computer Vision and Pattern Recognition (cs.CV); Robotics (cs.RO)
备注: 25 pages, 12 figures
Abstract:Vision-language-model-based embodied agents can complete instructed tasks but often violate safety constraints in the process, a problem recently framed as interactive safety. Training such agents to act safely is difficult, since safety and task success are distinct objectives, and safety arises only at a small number of safety-critical steps within a trajectory. Standard supervision is insufficient: imitating safe trajectories teaches behavior without explaining why it is safe, and contrasting arbitrary safe and unsafe trajectories mixes the safety signal with unrelated differences. We propose SafeBranch, a framework that aligns an embodied actor on safety through branch pairs constructed from the actor’s own unsafe rollouts via environment rollback. SafeBranch rolls each unsafe rollout back to the safety-critical step that caused the violation, queries the actor for a safe alternative, and pairs the original action with the alternative so that the two branches differ only at that step. The trained actor acts safely at deployment with no critic in the loop. On IS-Bench, SafetyALFRED, and out-of-distribution variants with unseen tasks and objects, it handles safety reliably without sacrificing task success, achieving roughly ten times more safe successes than the untrained baseline on the unseen-object variant.
[CV-44] Scale-Separated Conditioning for Style-Encoder-Free Diffusion Stylization
链接: https://arxiv.org/abs/2608.19719
作者: Jingtao Zhang,Haorui Gao,Youqing Liang,Zeming Liu
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
备注:
Abstract:Reference-based diffusion stylization requires separating target geometry from transferable appearance. Existing tuning-based methods often rely on aligned content-style-target triplets or auxiliary visual encoders, which increases data cost and can transfer unintended scene structure from the style reference. We propose SEFS (Style-Encoder-Free Stylization), a style-encoder-free conditioning framework for diffusion transformers. SEFS forms style tokens from stochastic low-resolution crops of single training images. This crop bottleneck preserves local appearance statistics such as palette, stroke, texture, and material, while reducing access to global layout cues. Target content is encoded by edge and segmentation cues and fused with the noisy latent through parameter-efficient trainable projections. We add style-to-denoising re-normalization for token-statistic alignment and cross-block skip fusion for spatial detail. SEFS trains on unpaired single images; the frozen diffusion VAE is used only to place image conditions in the latent space. On artistic stylization benchmarks, SEFS improves content consistency and leakage diagnostics while retaining reference-style affinity, and ablations support the crop-resolution, re-normalization, and skip-fusion choices. The code of SEFS will be made publicly available.
[CV-45] Robust Cross-Modal Foundation Model Perception for Underwater Robots under Degraded Visual Conditions
链接: https://arxiv.org/abs/2608.19710
作者: Mohammad Arif Ul Alam
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
备注:
Abstract:Reliable underwater robotic perception remains difficult because optical imagery degrades under turbidity, wavelength-dependent attenuation, low illumination, scattering, and blur. Although sonar provides complementary information that is less affected by optical visibility, prior visual-sonar research has largely focused on feature alignment and nominal detection performance. We investigate cross-modal robustness as visual reliability deteriorates and assess whether pretrained visual foundation-model representations can be complemented by sonar under severe degradation. We use frozen DINOv2 as the visual encoder and construct a controlled five-level benchmark ranging from clean to extreme visual conditions. We compare conventional visual detection, frozen foundation-model representations, sonar context, fixed multimodal fusion, clean-trained adaptive gating, and degradation-aware gated fusion. Our method trains the fusion mechanism across the full range of degradation while keeping the visual and sonar encoders frozen, allowing modality contributions to adapt without fine-tuning the pretrained backbone. Under extreme combined degradation, the DINOv2 baseline achieves 0.4610 balanced accuracy, while degradation-aware visual-sonar fusion reaches 0.6152, a 33.5% relative improvement. The learned sonar contribution increases from 14.2% under clean conditions to 41.3% under extreme degradation, demonstrating adaptive redistribution of cross-modal reliance. Fusion provides the largest gains under severe turbidity and blur, whereas color attenuation alone yields little additional benefit. These results show that foundation-model representations remain valuable but insufficient under severe information loss, and that explicitly adapting fusion to modality reliability can improve robust underwater multimodal perception.
[CV-46] RIPE: Reinforced Keypoint Learning from Positive Pairs Only ECCV2026
链接: https://arxiv.org/abs/2608.19693
作者: Johannes Künzel,Peter Eisert,Anna Hilsmann
类目: Computer Vision and Pattern Recognition (cs.CV); Machine Learning (cs.LG)
备注: LIMIT@ECCV 2026
Abstract:Sparse keypoint extraction and matching underpin core tasks in geometric computer vision, including structure-from-motion, visual SLAM, augmented reality, and medical image registration. Learning robust local feature representations, however, typically requires accurate camera poses or depth supervision, which are often unavailable in real-world settings. Reinforcement learning (RL) has recently emerged as a promising alternative, requiring only the information if two images show the same scene or not. However, existing RL formulations such as RIPE rely on coarse binary rewards and carefully constructed negative training pairs, limiting training stability and descriptor discriminability. In this paper, we revisit RL-based keypoint learning and propose a reward that fully exploits the geometric consistency signal, deriving both reward and penalty from a single positive pair without contrasting against negatives. This richer signal provides sufficient supervisory contrast to learn discriminative detectors and descriptors from positive image pairs alone, enabling representation learning under extremely limited supervision. Furthermore, we show that the same RL objective can be extended to the matching stage by adapting LightGlue, raising AUC@5 on MegaDepth1500 from 56.58 to 59.65 and enabling weakly-supervised training of the full sparse matching pipeline from image pairs with partial visual overlap. We validate our approach on established benchmarks, demonstrating competitive results compared to fully-supervised methods. We further show that the method can be even trained on low texture medical video sequences, where camera poses are usually unavailable and standard SfM pipelines often fail. Code and data are available at this https URL .
[CV-47] Scaffolding Minds: Optimizing Latent Visual Target Representations for Multimodal Reasoning
链接: https://arxiv.org/abs/2608.19669
作者: Haoqiang Kang,Yinpeng Chen,Luyang Liu,Jesper Sparre Andersen,Abhijit Ogale,Baochen Sun,Lichan Hong,Ed H. Chi
类目: Computer Vision and Pattern Recognition (cs.CV); Machine Learning (cs.LG)
备注:
Abstract:Latent reasoning has advanced multimodal reasoning through a two-stage training paradigm: (1) a helper image is encoded into latent tokens to teach visual chain-of-thought during a supervised fine-tuning (SFT) stage, and (2) these latent tokens are further refined with reward feedback during a reinforcement learning (RL) stage. In this paper, we identify two key limitations of this framework, one in each stage. First, the SFT stage typically relies on an off-the-shelf vision encoder to encode the helper image, yielding suboptimal latent representations that may not be well aligned with the downstream reasoning task. Second, existing RL methods treat the latent component only through deterministic regularization, which constrains policy drift but does not create alternative latent trajectories for exploration. To address these limitations, we propose Scaffolding Minds. Our approach learns a dedicated scaffolding encoder that provides an optimized target in latent space, and learns both the mean and variance of the RL sampler. We further show that these two improvements are complementary, together yielding substantial gains over strong baselines. Empirically, our method improves over the strongest latent-reasoning baseline by +9.5% on FrozenLake spatial planning, with the gain widening to +19% at 32x32 grid map, and by +5.2% on average across nine visual-centric reasoning benchmarks.
[CV-48] MUST-PET: MUltimodal Self-supervised learning across Tracers for whole-body PET/CT-based lesion segmentation
链接: https://arxiv.org/abs/2608.19666
作者: Bashirul Azam Biswas,Amartya Bhattacharya,Biratal Raj Wagle,Matthew E. Maeder,James B. Yu,Indrani Bhattacharya
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: Submitted to SPIE CAD 2027
Abstract:Deep learning-based whole-body PET-CT lesion segmentation can support cancer staging, treatment planning, and response assessment, but generalization is limited by scarce annotations and domain shifts. Self-supervised learning (SSL) can address these challenges but remains underexplored in pan-cancer, multi-tracer PET-CT. In this work, we propose MUST-PET (MUltimodal Self-Supervised learning across Tracers), a multimodal, multi-tracer SSL framework for generalizable whole-body PET-CT lesion segmentation. MUST-PET is trained and validated on a diverse, multi-institutional collection of pan-cancer PET-CT scans acquired with FDG and prostate-specific membrane antigen (PSMA)-targeted radiotracers. MUST-PET uses context-aware masked reconstruction, where one modality is partially masked and reconstructed using complementary information from both PET and CT. The pretrained model is subsequently fine-tuned with labeled samples and evaluated for reconstruction quality, lesion segmentation, label efficiency, and generalizability across independent held-out datasets. MUST-PET reduces reconstruction error, improves lesion segmentation over training from scratch, and performs well with limited labeled data and on unseen external datasets, demonstrating the potential of multi-tracer SSL for label-efficient, generalizable whole-body PET-CT. segmentation.
[CV-49] PL-NBA: A Possession-level Universal Basketball Video Dataset Supporting Multiple Visual Understanding Tasks
链接: https://arxiv.org/abs/2608.19646
作者: Yunhao Zhao,Haoying Sun,Jiarui Li,Zhuming Wang,Ya Jing,Xiangbo Shu,Lifang Wu,Changwen Chen
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Visual understanding in sports has emerged as a hot topic in computer vision in recent years. Most existing basketball video datasets adopt single action or activity as sample, which can neither preserve the temporal continuity of game events nor support complex tasks such as action anticipation. To address this issue, this paper constructs the first possession-level basketball video dataset (PL-NBA), in which each sample is composed of a complete NBA offensive possession. Collected from 60 NBA games, PL-NBA contains 11,000 valid offensive possession clips and 31,567 annotated events with player names, captions, event types and timestamps. Each video clip includes multiple events and preserves the continuity of events, which is helpful for analysis of tactic. Experiment is conducted on multiple visual understanding tasks, including event recognition, video captioning, temporal action localization and action anticipation. Experimental results show that existing methods achieve limited performance on above four tasks, demonstrating that PL-NBA is a challenging benchmark for sports video understanding.
[CV-50] When Guidance Goes Off-Scale: Recalibrating Diffusion Transformers under Analog Compute-in-Memory Nonidealities
链接: https://arxiv.org/abs/2608.19644
作者: Wenshuai Yao,Wenyong Zhou
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: 9 pages, 8 figures, 3 tables
Abstract:Diffusion Transformers (DiTs) incur high memory traffic and energy costs because sampling repeatedly evaluates large denoisers dominated by linear operations. Analog compute-in-memory (CIM) can alleviate these costs by executing linear operations within weight-storing memory arrays. However, CIM nonidealities perturb effective weights, with errors accumulating along the state-dependent denoising trajectory; their interaction with classifier-free guidance (CFG) remains underexplored. In this paper, we characterize the impact of analog CIM nonidealities on DiT sampling. Although conditional and unconditional predictions can each remain close to their clean counterparts, their difference (the CFG residual) is disproportionately attenuated and rotated. Identifying this residual as a controllable failure channel, we propose a retraining-free, sampler-side recalibration that adjusts only the CFG scale for a given CIM condition. Trajectory-level analysis shows that moderate recalibration strengthens the target-oriented component preserved in the distorted residual, enabling earlier commitment to a prompt-consistent semantic region. In contrast, excessive guidance amplifies the full noisy residual and degrades quality, resulting in a finite, noise-dependent optimum. Extensive experiments on PixArt-Sigma, PixArt-alpha, and DiT-XL/2 show that the optimal guidance scale increases with CIM noise. Using 30,000 samples per condition, guidance recalibration consistently restores generation quality across simulated CIM mappings, closing at least 87% of the CIM-induced FID gap at a CIM noise level of 0.20. It reduces FID from 59.22 to 20.49 on PixArt-Sigma, 72.37 to 21.12 on PixArt-alpha, and 20.89 to 6.62 on DiT-XL/2.
[CV-51] S2GS: Structured Sparse Gaussian Streaming for Efficient Free-Viewpoint Video Reconstruction on Edge-IoT Devices
链接: https://arxiv.org/abs/2608.19639
作者: Yiwei Li,Jiannong Cao,Weixun Gao,Rui Cao,Songye Zhu,Yinfeng Cao,Mingjin Zhang
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: Project Page, Code, and Supplementary Material: this https URL
Abstract:Streaming reconstruction of Free-Viewpoint Videos (FVVs) supports immersive Internet of Things (IoT) services, such as telepresence and digital twin visualization. Existing methods suffer from high per-frame optimization time and large storage footprints, limiting deployment on resource-constrained Edge-IoT devices. To address these challenges, we propose Structured Sparse Gaussian Streaming (S ^2 GS), an FVV reconstruction framework that exploits structure-aware temporal sparsity to selectively update Gaussian residuals, enabling efficient streaming without compromising visual fidelity. In the spatial domain, a streaming octree hierarchically organizes Gaussian residuals, capturing spatial correlations that guide residual updates. In the temporal domain, a structured gating mechanism, comprising hierarchical feature propagation (HFP) and Gumbel-Sigmoid sampling, converts hierarchical dynamic cues into sparse residual update decisions under differentiable optimization. A multi-level discrete scheme is further adopted to provide fine-grained control over residual updates while preserving intricate dynamic details. Extensive experiments across consumer GPUs, industrial edge IoT devices, and a physical telepresence testbed demonstrate that S ^2 GS consistently reduces per-frame optimization time and storage footprint while maintaining competitive visual quality. Compared with QUEEN, S ^2 GS reduces per-frame optimization time by 59% and storage costs by 85% on an RTX 4090 GPU. On the Jetson AGX Orin, S ^2 GS delivers the highest rendering throughput (60+ FPS) and the lowest energy consumption among the evaluated methods, demonstrating its potential for deployment in resource-constrained systems.
[CV-52] xtRefine: Improving Textual Fidelity Spatial Placement and Glyph Rendering for Text Editing in Product Posters
链接: https://arxiv.org/abs/2608.19637
作者: Honglie Wang,Jia Sun,Zijun Li,Junlong Wu,Pengcheng Wei,Jiyuan Wang,Yongrui Heng,Boheng Zhang,Huaiqing Wang,Dewen Fan,Qianqian Gan,Fan Yang,Tingting Gao,Yan-Ming Zhang
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Text editing in product posters entails inserting new text or replacing existing text while preserving product appearance, background content, and global composition. Despite recent progress in instruction-based image editing, general-purpose models remain unreliable in this setting: they often omit or incorrectly render the target text, place it over salient products or pre-existing content, and produce structurally distorted or visually inconsistent glyphs. We introduce \textbfTextRefine, a task-aligned post-training framework that combines supervised fine-tuning with operation-specific reward optimization to address these complementary failure modes. For text insertion, our text-span-level reward jointly assesses semantic fidelity and target-span coverage, penalizes spatial conflicts with products and existing text, and employs a gated structural constraint to preserve non-text regions. For text replacement, our glyph-level reward leverages the connectionist temporal classification (CTC) posterior of the target character to provide graded supervision for fine-grained defects, including missing strokes, structural deformations, and confusion among visually similar characters. We further introduce \textbfOpenTextEdit, a dataset comprising 100K images for text editing in product posters, with multi-text layouts, detailed text attributes, product masks, and challenging low-frequency characters. Extensive experiments on both insertion and replacement demonstrate that TextRefine consistently outperforms the evaluated image editing baselines in textual fidelity, placement reliability, and glyph quality while better preserving source-image content.
[CV-53] What Matters for Latent Actions in Robot Learning
链接: https://arxiv.org/abs/2608.19613
作者: Xizhou Bu,Qingda Hu,Lei Zhou,Lingfeng Zhang,Yingbo Tang,Zihao Liu,Xinyi Tao,Zhiqiang Ma,Qingqiu Huang,Chufeng Tang,Hongbo Wang,Jing Zhang,Jiayi Ma,Hangjun Ye,Wei Li,Xiaoshuai Hao
类目: Robotics (cs.RO); Computer Vision and Pattern Recognition (cs.CV)
备注: Project page: this https URL
Abstract:Latent Action Models (LAMs) have emerged as a promising paradigm for enabling robot learning to leverage large-scale unlabeled videos through latent actions that serve as compact surrogates for physical actions. Despite rapid progress, research on LAM remains highly fragmented, with existing methods evaluating different design choices in isolation under inconsistent experimental settings, making it difficult to identify the factors that truly determine downstream robotic manipulation performance. In this work, we present the first comprehensive empirical study of latent action learning for robotic manipulation. We unify representative LAM methods within a common autoencoding framework and systematically investigate 41 LAM design choices across three dimensions, including latent action modeling paradigms, learning objectives and regularization methods, and latent action integration strategies. We further examine four proxy metrics for evaluating latent action quality and assess their ability to reliably predict downstream robotic manipulation performance. Extensive experiments on three widely used benchmarks provide strong empirical evidence that fine-tuning vision-language model (VLM) backbones with latent actions provides a stronger initialization for downstream policy learning, with further validation on real-world robot manipulation tasks.
[CV-54] OrthoSkillVLA: Continual Skill Learning via Gradient-Informed Skill Subspace Adaptation
链接: https://arxiv.org/abs/2608.19589
作者: Jiaqi Wang,Zhou Fang,Qiongfeng Shi,Yi Zhou
类目: Robotics (cs.RO); Computer Vision and Pattern Recognition (cs.CV)
备注: Accepted by PRCV 2026
Abstract:Pretrained Vision-Language-Action models provide a strong foundation for robot learning, but sequentially adapting them to diverse skills can perturb the representations and velocity mappings used by previous skills, leading to catastrophic forgetting. Architecture-based approaches improve retention by isolating skills but lead to increased inference footprint. Recent subspace-constrained methods restrict parameter updates in an orthogonal subspace to minimize interference but impose a unified constraint on the entire model. We analyze the distinct roles of internal VLA components and identify two VLA-specific challenges. First, the VLM maintains broad semantic representations, making it vulnerable to capacity exhaustion, whereas the ActionHead refines semantics into localized velocity patterns that are highly sensitive to perturbations. Second, the final velocity decoder serves as a readout layer. Freezing it forms an output-stage expressivity bottleneck, while updating it risks overwriting previous velocity mappings. To this end, we propose OrthoSkillVLA, a parameter-efficient framework for continual skill learning in pretrained VLA models without demonstration replay. Given the representation heterogeneity, we impose separate subspace constraints on the VLM and ActionHead, preserving reusable semantic capacity while protecting localized velocity patterns. For the output layer, we introduce a lightweight feature-aware MoE decoder, where each skill is allocated a compact expert and a training-free router selects the expert according to feature-space affinity. Extensive simulated and real-world evaluations, together with ablations, demonstrate that OrthoSkillVLA better preserves prior skills while acquiring new ones.
[CV-55] VGI-BENCH: Probing Visual Intelligence in Video Generation Models
链接: https://arxiv.org/abs/2608.19583
作者: Xuan He,Cong Wei,Yuhao Cheng,Linrui Ma,Yuxuan Zhang,Zuojun Li,Yuhao Wen,Zeyi Liu,Yuren Hao,Songcheng Cai,Keming Wu,Penghui Du,Kai Zou,Rui Yang,Chenkai Sun,Ke Yang,Ping Nie,Kelsey R Allen,Chenglong Wang,Michel Galley,Jianfeng Gao,ChengXiang Zhai
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
备注:
Abstract:Recent studies suggest that video generation models can exhibit certain forms of zero-shot visual reasoning through generated frames. Yet reliable evaluation remains challenging: benchmarks should adopt inputs aligned with the visual priors of current video models, require valid evolving processes rather than only plausible final states, and calibrate task difficulty to remain challenging yet partly feasible. To this end, we introduce VGI-bench, containing 27 tasks and 810 instances, organized by a two-level taxonomy of task domains and skill tags for fine-grained evaluation of visual reasoning capabilities of video generation models. Our evaluations show that current generative systems can solve a subset of visually grounded reasoning tasks, but remain far from reliable, with even the strongest model, Seedance~2.0, achieving only 51.0% under our evaluation criteria. Our analysis further explore the output failure modes, input condition sensitivity, performance transfer boundary from synthetic fine-tuning, and internal denoising perspective revealing limited self-correction, where later steps mainly refine early hypotheses rather than correct reasoning errors. We hope VGI-bench will help stimulate the development of next-generation video generation models. We will release our code and data.
[CV-56] MixFix-Net: A Dual-Stage Trajectory Prediction Model for AIS and Vision-Derived Vessel Data
链接: https://arxiv.org/abs/2608.19580
作者: Md Mahmuddun Nabi Murad,Bora San Turgut,Yasin Yilmaz
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Vessel trajectory prediction is critical for maritime safety and accident prevention. While most existing trajectory prediction models rely on Automatic Identification System (AIS) data due to its precision and availability, small vessels mostly operate without AIS, resulting in a significant monitoring gap. To address this, we propose MixFix-Net, a dual-stage mixer-based trajectory prediction model designed to handle vessel trajectory time-series data derived from both AIS and (non-AIS) vision data. Our architecture integrates a Primary Trajectory Predictor with a Residual Trajectory Adjuster, enabling more refined trajectory prediction. Additionally, we introduce a new video-based dataset derived from webcam streams, from which vessel trajectories are extracted to represent non-AIS data. Extensive evaluations on both AIS and non-AIS datasets across six metrics (mean squared error, mean absolute error, symmetric mean absolute percentage error, final displacement error, Frechet distance, and average Euclidean distance) demonstrate that MixFix-Net consistently outperforms existing baselines across most metrics and datasets.
[CV-57] Block3D: Efficient Text-to-3D Generation via Block-Wise Diffusion
链接: https://arxiv.org/abs/2608.19567
作者: Bowen Cui,Weijie Wang,Zeyu Zhang,Yefei He,Mingda Lin,Haoyu Zhao,Yuanyu He,Donny Y. Chen,Feng Chen,Bohan Zhuang
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: Project page: this https URL
Abstract:While text-to-3D generation has advanced rapidly, achieving high geometric fidelity at low inference cost remains challenging. Existing text-to-3D methods either decode discrete shape tokens autoregressively or iteratively refine global 3D representations with diffusion or flow models. However, autoregressive decoding is sequential and cannot revise errors, whereas diffusion and flow-matching models repeatedly process the full representation, making high-quality generation increasingly expensive. In this paper, we propose Block3D, a block-wise diffusion framework that partitions the discrete shape-token sequence into contiguous blocks, generates the blocks autoregressively, and jointly denoises all tokens within the current block. To alleviate error accumulation, we introduce confidence-guided intra-block correction, which revises low-confidence tokens before each block is finalized. On a held-out set from TRELLIS-500K, Block3D reduces mean end-to-end generation time from 25.71 seconds to 4.99 seconds, achieving a 5.15\times speedup over the fine-tuned autoregressive baseline without sacrificing geometric fidelity.
[CV-58] Stream4D: 4D-Consistency for Streaming Autoregressive Diffusion Video Models
链接: https://arxiv.org/abs/2608.19556
作者: Yuanhao Ban,Jiaqi Feng,Hengguang Zhou,Xiaohuan Pei,Justin Cui,Cho-Jui Hsieh
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
备注:
Abstract:Streaming autoregressive diffusion models enable real-time, long-horizon video generation, but their training objectives optimize local frame prediction rather than the geometry and dynamics of a coherent world: long rollouts accumulate geometric drift and degrade into static or unnatural motion. Recent bidirectional approaches address this problem using rewards signals built upon 3D Gaussian-Splatting reconstruction. However, a single rigid 3d reconstruction cannot model a dynamic scene, so this critic penalizes genuine object motion as reconstruction error and is maximized by freezing the video. This shortcut is especially detrimental in the AR setting, where each chunk can propagate an already-static configuration. In this work, we propose Stream4D, which replaces the static critic with a feed-forward 4D reconstruction reward that explicitly models scene dynamics, allowing coherent motion to receive high consistency rewards. To further guide motion magnitude and quality, we add a motion prior that rewards natural scene-flow magnitude while penalizing jitter and non-rigid artifacts. Our final recipe combines these two terms with a lightweight perceptual anchor. Across various autoregressive video backbones and various generation horizons, Stream4D improves 4D reconstruction quality, preserves motion more effectively, and achieves higher human-aligned preference. Project page: this https URL
[CV-59] Where Grounding Accuracy Lives on the IoU Curve: Label-Free Inference-Time Boundary Refinement
链接: https://arxiv.org/abs/2608.19553
作者: Bo Ma
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Vision–language models can identify the correct referent while returning an imprecise bounding box. We study whether a frozen direct-answer model can use its own prediction to allocate one additional localized observation without accessing target annotations at inference. Label-free precision refinement (LFPR) routes predicted-small regions to a higher-resolution pass, re-grounds the expression inside a context crop, admits a candidate only under fixed geometric guards, and returns a fixed coordinate-wise midpoint. We report results across three evidence tiers. On 31,921 retrospective Ref-L4 expressions, LFPR raises mAcc _0.5:0.95 from 72.947% to 76.013% (Acc@0.5 88.531% \to 89.725%, Acc@0.9 55.788% \to 61.142%). A frozen transfer to 30,969 RefCOCO/RefCOCO+/RefCOCOg expressions improves every dataset at Acc@0.5, mAcc, and mean IoU (pooled mAcc +0.645 , Acc@0.5 +0.817 ), while Acc@0.9 is unchanged overall: routing alone gains +1.162 points there, but crop, guards, and fusion give back -1.192 , offsetting rather than showing no strict-IoU effect. A prospective, image-disjoint Flickr30K Entities evaluation improves every endpoint (mAcc +0.973 , Acc@0.9 +1.022 ), more strongly under a single-box variant (mAcc +2.575 , Acc@0.9 +3.689 ). The same operator applied to two released grounding specialists improves every endpoint (Acc@0.9 +1.569 / +6.716 for EGM-4B/8B) at roughly twice the latency, composing with specialist training rather than replacing it. A genuine unguarded control (guard removed from the same candidates) underperforms the incumbent on every metric, showing the guard is load-bearing. Together, these results show that referent selection and boundary precision are partially separable, with different components moving opposing regions of the IoU curve – behavior a single threshold cannot reveal.
[CV-60] Continuous Adversarial MeanFlow Transfer
链接: https://arxiv.org/abs/2608.19540
作者: Yara Bahram,Zahra Dehghani,Mélodie Desbos,Eric Granger,Pablo Piantanida,Mohammadhadi Shateri
类目: Machine Learning (cs.LG); Computer Vision and Pattern Recognition (cs.CV)
备注: Paper under review
Abstract:Training fast generators on new domains with limited data remains challenging for two reasons. First, adapting a pretrained diffusion or flow model to a new domain leaves its costly multi-step sampling unaddressed, and existing acceleration methods are tied to the source parameterization-- \epsilon , x , v , or u --leaving heterogeneous pretrained models with no common acceleration target. Second, while adversarial refinement is proven effective for few-step quality, it is formulated only for instantaneous-velocity flows, not for the finite-interval average velocities that MeanFlow (MF) models predict. We address both problems. We propose MeanFlow-Transfer, which maps heterogeneous source outputs into a shared velocity representation, uses it to initialize an MF generator from the source weights, and optimizes an MF objective on the target domain. This unifies adaptation and acceleration in a single training loop across a broad range of pretrained models. We then introduce Continuous Adversarial MeanFlow, a post-training stage that extends continuous adversarial flow models from instantaneous velocities to MF’s finite-interval average velocities. CAMF contrasts changes in a learned potential between real and predicted interval endpoints, recovering fine detail that MF regression averages away, and reduces to the instantaneous criterion in the vanishing-interval limit. Adapting four ImageNet-based source models–DiT ( \epsilon ), SiT ( v ), JiT ( x ), iMF ( u )–to five target domains, MF-T with CAMF matches or exceeds the fine-tuned teacher in FID and FDD at up to 125\times fewer Neural Function Evaluations (NFEs), while CAMF improves MF-T’s few-step FID by 29% on average.
[CV-61] CVSD-Reg: Cross-Modal Visual Semantic Prior Distillation for Robust LiDAR Registration
链接: https://arxiv.org/abs/2608.19536
作者: Eunsoo Im,Junghun Suh,Gyeonggwan Lee,Seunghwan Hong
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI); Robotics (cs.RO)
备注:
Abstract:Learning-based global point cloud registration has achieved remarkable progress, yet its reliance on geometric representations makes existing methods sensitive to variations in point density, scan pattern, viewpoint, and sensor characteristics. We propose CVSD-Reg, a robust global LiDAR registration framework that distills visual semantic priors from a vision foundation model into LiDAR representations. In Stage 1, a Point Transformer V3 student learns from a frozen DINOv2 teacher through contrastive distillation and spherical-manifold alignment, which preserves the hyperspherical geometry of the teacher embedding space. Self-supervised InfoNCE consistency and soft \mathrmSE(3) invariance further encourage viewpoint-robust descriptors. In Stage 2, the distilled representation is adapted to registration through correspondence learning, density-aware point-dropout augmentation, and end-to-end pose optimization. With a single checkpoint, CVSD-Reg generalizes to both single-sensor and zero-shot cross-sensor scenarios without sensor-specific adaptation and remains entirely camera-free at inference. On KITTI, nuScenes, and HeLiPR, CVSD-Reg achieves strict success rate (SR@0.5,m/ 1^\circ ) of 97.7 % , 99.0 % , and 99.3 % , respectively, including 97.3 % on sparse 16-beam Velodyne scans. It outperforms state-of-the-art geometric registration methods by up to 44.0 percentage points without requiring camera inputs or post-hoc ICP refinement.
[CV-62] LF-GICP: Parameter-Free Degeneracy-Aware LiDAR Odometry via a Voxel-Normal Localizability Field
链接: https://arxiv.org/abs/2608.19522
作者: Eunsoo Im
类目: Robotics (cs.RO); Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Scan-to-map LiDAR odometry drifts unboundedly along the unobservable axes of geometrically degenerate environments like tunnels and corridors, and existing degeneracy handling requires environment-specific parameter tuning. This paper presents a parameter-free approach. We show that in voxelized GICP the Gauss–Newton (GN) Hessian masks translational degeneracy, because covariance regularization keeps the translation block artificially well-conditioned. We bypass this with a regularization-free voxel-normal localizability field and two of its statistics: a normalized fraction f_0 detecting directional anisotropy, and an absolute per-voxel mass \lambda_0 distinguishing information absence (tunnels) from dilution (dense open scenes). A temporal-median gate combines both to trigger Fisher-information correspondence weighting. Calibrated once by fixed rules on two short sequences and then frozen, LF-GICP achieves the lowest KITTI relative translation error ( 0.865% ) under an identical evaluation protocol against re-run baselines, outperforms them on GEODE tunnels and MulRan, leads the HeLiPR mean, and generalizes across four sensor types without re-tuning. We further demonstrate empirically that straight, uniform tunnels remain unobservable along their axis for LiDAR-only registration.
[CV-63] A Plug-in Interpretation of Conditioning in Score-Based Diffusion Models BMVC2026
链接: https://arxiv.org/abs/2608.19504
作者: Libo Chen,Souvik Ghosh,Teo Deveney,Chris Budd,Vinay P. Namboodiri
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: Accepted at BMVC 2026
Abstract:We propose a conditioning mechanism for diffusion models based on multi-speed joint diffusion of the target and the condition. The mechanism learns an unconditional joint score network and enforces conditioning at inference via a plug-in correction term. The plug-in term separates the conditioning contribution from the learned unconditional dynamics, offering a transparent view of how the condition steers generation of the target distribution. Building on this, we derive explicit conditional reverse-time SDEs and approximate probability-flow ODEs, enabling principled and directly comparable conditional samplers. To reduce the induced ODE–SDE discrepancy, we introduce a log-Fokker–Planck residual regularization that improves ODE sampling quality. Experiments on conditional image generation tasks demonstrate competitive performance and support the effectiveness of the plug-in conditioning view. Additional ODE–SDE comparison experiments show that the log-Fokker–Planck residual regularization improves deterministic ODE sampling.
[CV-64] Fine-Tuning VLAs with Self-Demonstrated Generative Control for Multi-Task Manipulation
链接: https://arxiv.org/abs/2608.19490
作者: Prachi Garg,Steve Xing,Prahit Yaugand,Saurabh Gupta,Derek Hoiem
类目: Robotics (cs.RO); Computer Vision and Pattern Recognition (cs.CV); Machine Learning (cs.LG)
备注: Project Page: this https URL
Abstract:State-of-the-art vision-language-action (VLA) models such as \pi_0.5 exhibit strong semantic understanding, instruction following and task behavior. However, when deployed on new robots, even minor mismatches in hardware configuration relative to pretraining can cause severe performance drops. Finetuning the VLA on in-domain expert data from the new embodiment improves performance on the expert task but leads to a loss in its original instruction following and behavioral priors. In this paper, we propose a self-supervised method that generates online interaction rollouts from the zero-shot VLA as additional training data for finetuning. Our experiments show this finetuning scheme yields strong multi-task policies that, on the target robot, (1) inherit prior tasks distilled from the zero-shot model, (2) enable generalist instruction following, while (3) learning new skills from expert data with improved sample efficiency. We demonstrate the success of our approach across test sets probing generalization on a real ALOHA robot and a new simulation benchmark in RoboTwin. Video results are available at this https URL
[CV-65] VideoRun2D Demo: Markerless Body Tracking for Biomechanical Analysis of Running CVPR
链接: https://arxiv.org/abs/2608.19480
作者: Luis F. Gomez,Julian Fierrez,Roberto Daza,Ruben Tolosana,Aythami Morales,Gonzalo Garrido,Javier Rueda,Enrique Navarro
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: 5 pages, 4 figures, 2 tables. IEEE/CVF Conf. on Computer Vision and Pattern Recognition Workshops (CVPRW), 2026 (1st PhysHuman Workshop: Physically Grounded Human Perception and Modeling)
Abstract:Human pose estimation has advanced significantly due to the development of deep learning models, increased data availability, and improved computing resources. These developments have led to highly accurate body tracking systems with direct applications in sports analysis and performance evaluation. The VideoRun2D Demo performs a biomechanical analysis during sprints using different human pose estimators. The proposed framework was evaluated using human pose trackers and expert manual annotations. The tested framework uses 314 sprints from 44 professional runners, focusing on two key joint angles in sprint biomechanics: 1) hip flexion/extension and 2) knee flexion/extension. The framework also includes a post-processing module for outlier detection. The tested results demonstrate that the average root-mean-square errors range from 11.46° to 5.83° for the best trackers. When integrated with the post-processing modules, these errors can be reduced to 9.87° and 5.30°, respectively. The VideoRun2D Demo findings suggest that human pose-tracking approaches can be valuable resources for the biomechanical analysis of running.
[CV-66] HiRA-CAM: Preserving Fine-Grained Spatial Relevance in Gradient-Based Visual Explanations
链接: https://arxiv.org/abs/2608.19407
作者: Manasi Nerurkar,Ali A. Minai
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI); Machine Learning (cs.LG); Neural and Evolutionary Computing (cs.NE)
备注: IEEE World Congress on Computational Intelligence, Maastricht, Netherlands, June 2026
Abstract:Deep Learning models can include billions of parameters or more, making it difficult to explain their internal transformations and outputs. However, explainability is increasing in importance due to the use of AI in crucial applications. This paper focuses on the interpretability of convolutional neural networks (CNNs). Building on the popular gradient based method LayerCAM for extracting internal features in CNNs, we propose an improved method named HiRA-CAM, and show that it outperforms both LayerCAM and Grad-CAM on creating useful saliency maps for object classification. The main feature of HiRA-CAM is its adaptive use of activation maps from all the layers of the CNN to arrive at a more focused saliency map.
[CV-67] Beyond Recognition: Compact Multi-Domain Arabic Manuscript HTR with Candidate-Selection Analysis and Evidence-Preserving Review
链接: https://arxiv.org/abs/2608.19385
作者: Abdullah Ahmed Ali,Mohammed Thamer Abdulhadi,Ali Haider Safaa,Dhulfiqar Mahdi Wadi
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: 13 pages, 4 figures, 12 tables
Abstract:Historical Arabic manuscript transcription is not only a recognition problem. A usable scholarly system must cope with shifting hands and layouts, preserve uncertain readings, distinguish visual evidence from linguistic plausibility, and record the researcher’s final decision. We present Phoenix, a 4.99-million-parameter CNN-BiLSTM-CTC recognizer, and Athar, an evidence-aware review workflow built around it. Phoenix is adapted across archival, Maghrebi, and historical manuscript domains using document-aware replay, an expanded 81-symbol codec, and forgetting guards that reject checkpoints that improve a new domain at unacceptable cost to previous domains. In a pre-specified held-out comparison against the preceding checkpoint, frozen before evaluation and scored with greedy decoding and raw references, Phoenix reduced CER from 22.12% to 17.86% on 10,594 Agapet lines and from 17.72% to 11.84% on 11,684 Omar lines, while regressing from 10.39% to 10.72% on 164 TariMa lines. Across the two large held-out sets, character-weighted CER fell from 19.98% to 14.93%, a 25.3% relative error reduction. A separate same-protocol development diagnostic found the lowest CER for Phoenix on four of four comparable domains (9.59% unweighted macro CER). An N-best diagnostic revealed a 2.15-point oracle gap between beam decoding and Oracle@25, while neural text rerankers, consensus MBR, CTC-posterior quality estimation, and local pre-CTC hidden-state quality estimation recovered less than 4% of this gap. Athar therefore preserves the visual reading, exposes bounded alternatives, uses local language models conservatively, retrieves source parallels with unique, ambiguous, or abstain states, and exports auditable TEI and PAGE-XML records. The results support evaluating manuscript HTR as auditable evidence management rather than silent text replacement.
[CV-68] CAViAR: A Causal Video Dataset for Fine-Grained Accident Reasoning in Real-World Scenarios ECCV2026
链接: https://arxiv.org/abs/2608.19380
作者: Sparsh Garg,Yi-Wen Chen,Vijay Kumar B G,Abhishek Aich
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: Accepted to ECCV 2026 Workshop DriveX
Abstract:While modern autonomous driving systems excel at perception tasks such as object detection and trajectory prediction, they lack the high-level causal reasoning required to interpret traffic accidents. In particular, determining responsibility, such as identifying who is at fault and which traffic rule was violated, remains largely unexplored in current benchmarks. To this end, we introduce CAViAR (Causal Accident Video and Incident Analysis Repository), a human-annotated dashcam benchmark comprising 2,249 real-world accident videos collected from CarCrashDataset (CCD) and Nexar. Each video is annotated with structured labels spanning environmental conditions, accident type, causal explanation, apparent At-Fault Agent, affected agent, and apparent rule-violation category. We benchmark state-of-the-art vision-language models (VLMs), including Cosmos-Reason2, Qwen3-VL, and InternVL3. Once class imbalance is accounted for with majority/random baselines and balanced metrics, perceptual competence is uneven–lighting is nearly solved, whereas weather and road-condition accuracy fall at or below the majority-class baseline—and all models degrade sharply on accident type and responsibility reasoning. Overall, CAViAR exposes a practical Perception–Reasoning Gap: current VLMs may recognize salient context, but do not reliably map visible agent actions to annotated rule-relevant responsibility categories in safety-critical driving scenarios. Code, annotation schema, prompts, and evaluation scripts are available at: this https URL
[CV-69] Does Marginal Coverag e Guarantee Class-Conditional Safety for Zero-Shot VLMs Under Shift? ECCV2026
链接: https://arxiv.org/abs/2608.19376
作者: Jai Kumar Sharma,Amartya Dutta
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
备注: Accepted at the ECCV 2026 Workshop on Uncertainty Quantification for Computer Vision (UNCV). 34 pages (16 main + 18 supplementary), 10 figures
Abstract:Split-conformal prediction provides marginal coverage under exchangeability and is increasingly used as an abstention layer for zero-shot vision-language models (VLMs). We audit this practice under deployment shift for CLIP, OpenCLIP, and SigLIP across ImageNet and non-ImageNet settings. Marginal coverage can remain relatively high while class-conditional tail coverage collapses: on ImageNet-Sketch, worst-class coverage falls to \approx 0 and 10-12% of classes lie below a finite-sample null floor, despite marginal coverage of about 0.86. The failure is aligned with target-domain class accuracy but is not predicted by the source-domain diagnostics we test. Source-side Mondrian calibration improves the in-distribution tail but does not transfer, while clustered conformal and Conf-OT improve marginal or average metrics without recovering the worst-class tail. Target-side class calibration substantially lifts the tail, but requires labels for every class and remains set-size-intensive. We further identify a 2-3 \times cross-family efficiency gap and show that native SigLIP sigmoid scores remove APS’s probability-mass interpretation. The findings persist across the tested model scale, pretraining corpus, prompt, miscoverage level \alpha , and shifted non-ImageNet settings. Marginal conformal coverage should therefore be treated as an average reliability statistic, not as a safety guarantee for the class tail.
[CV-70] GRACE: Grounded Reasoning via Adapter Composition and Evidence-Aware Calibration for Educational Visual Question Answering
链接: https://arxiv.org/abs/2608.19355
作者: Xinjin Li,Yudi Xia,Xi Zhao,Yiliu Xu,Yining Liu,Cheng Lu,Yujian Long,Yu Ma,Jinghan Cao,Liang Fan,Yeyun Xu
类目: Multimedia (cs.MM); Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Educational visual question answering, or VQA, requires models to solve curriculum-oriented multiple-choice questions using both language and visual evidence. Compared with conventional open-ended VQA, educational examples often include structured assessment metadata, diagrams or image contexts, and semantically close answer options, creating strong opportunities for question-option shortcuts. We develop and evaluate a parameter-efficient adaptation framework for a frozen multimodal large language model in this setting. We introduce GRACE, Grounded Reasoning via Adapter Composition and Evidence-Aware Calibration, a framework that uses the pedagogical state of each question to specialize lightweight language and vision adaptation. The state combines inference-visible subject, grouped skill, grade, visual-context, question-intent, and option-structure cues. GRACE uses factor-specific prompts and lightweight visual adapters, then applies evidence-aware option calibration to score all candidates under a shared multimodal context. On ScienceQA, GRACE improves a shared-adapter baseline from 90.5 percent to 93.1 percent overall accuracy and from 88.7 percent to 91.2 percent on image-context questions. Removing pedagogical composition, option calibration, or the visual adapter reduces overall accuracy by 1.4, 1.0, and 1.5 points, respectively. These controlled results show that structured educational state is an effective routing signal for parameter-efficient multimodal adaptation.
[CV-71] SceneGTMM: A Conformal Mapping-based Scene-Aware Transferable GNN-Transformer Dual-Graph Interaction Framework for Map Matching
链接: https://arxiv.org/abs/2608.19298
作者: Yongliang Zhang,Feng Song,Ji Chen,Lishuai Guo,Yong Deng,Yue Zheng,Tianyi Liu,Zhixiong Chen,Qixin Zhang
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Map matching is a key technology connecting positioning data with high precision road networks, but it faces challenges in noise robustness, cross regional transfer, and interpretability. To addr ess the limitations of existing methods in local global fusion, dynamic road network adaptation, and reliance on black box mod els, this paper proposes SceneGTMM, a transferable GNN Transformer dual graph interaction map matching framework based on a conformal mapping based scene relative strategy. 1) Conformal mapping based scene relative strategy: constructs trajectory centric local coordinate systems to reduce dependence on the training road network, supporting cross regional transfer and dynamic road network updates; 2) GNN Transformer dual graph interaction architecture: a GNN modeled road graph captures local topological constraints, while a Transformer modeled trajectory graph captures global temporal dependencies, and cross graph attention achieves noise suppression and semantic alignment; 3) CRF enhanced structured prediction: combines the global context of the Transformer with the topological transition constraints of CRF to improve path connectivity and robustness. Experiments show that SceneGTM achieves over 80% accuracy on multi source trajectories with positioning errors of 16 50 meters, representing a 5.3% improvement over HMM. In cross city transfer scenarios, it outperforms MTrajRec, GraphMM, and TMM, and enhances interpretability through attention and relative coordinate visualization. This study provides a new paradigm for high precision, transferable map matching for real time traffic perception and autonomous driving path planning.
[CV-72] Clustering and Token Denoising for Faster and More Robust VLMs
链接: https://arxiv.org/abs/2608.19285
作者: Baptiste Rossigneux,Inna Kucher,Vincent Lorrain,Emmanuel Casseau
类目: Computer Vision and Pattern Recognition (cs.CV); Machine Learning (cs.LG)
备注:
Abstract:Recent Visual-Language Models (VLMs) have enhanced the capabilities of pre-trained LLMs by adding vision tokens alongside text, with approaches like LLaVA showing impressive results. However, the computational burden of processing up to 576 or 729 visual tokens makes edge deployment challenging. While various token pruning techniques require retraining, some are training-free and thus can easily adapt to architecture changes. We introduce ClustRS, a two-part, training-free algorithm for robust token pruning. Its first component is an attention-weighted, clustering algorithm that selects representative tokens from each semantic cluster. The second component, Residual Shrinkage, is a one-pass denoising step on the selected tokens. These training-free lightweight steps make LLaVA ready for real-world data, improving robustness to a wide range of image-noise types and intensities. Experimental results on the ScienceQA-IMG and MM-VET benchmarks show our method outperforms attention- and diversity-based methods by up to 20% under extreme noise and token conditions (reducing tokens by 97%, down to 16 tokens) on LLaVA 1.5 7b and achieves exceptional results on LLaVA-OneVision, where we match baseline performance with fewer than one-third of their tokens under mild noise conditions. Our study demonstrates a simple yet powerful alternative to both score-only and diversity-only pruning rules, paving the way for compute-efficient and noise-resilient VLM deployment.
[CV-73] Spiking Local Interaction and Adaptive Complementary Fusion for Spiking Transformer
链接: https://arxiv.org/abs/2608.19238
作者: Dongcheng Zhao,Sicheng Shen,Zhenyu Yang,Zhiyuan Li,Jinyan Yu,Yongjian Wang,Tiechui Yao,Wenli Zhang,Tielin Zhang
类目: Neural and Evolutionary Computing (cs.NE); Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Spiking Transformers model token interactions primarily through spiking self-attention (SSA). However, binary query and key representations map continuous similarities to sparse and discrete relation responses, which may suppress weak relations and limit the propagation of local spatial context. To address this limitation, we introduce Spiking Local Interaction (SLI) and Adaptive Complementary Fusion (ACF). SLI establishes an attention-independent pathway for direct information exchange among neighboring spiking tokens using lightweight depthwise–pointwise transformations. ACF integrates SSA and SLI through layer-specific, channel-wise coefficients that adaptively balance their contributions at different network depths. The proposed design preserves the original attention formulation and can be incorporated into different Spiking Transformer architectures with modest parameter overhead. Experiments on ImageNet-1K, CIFAR-10, CIFAR-100, CIFAR10-DVS, and ADE20K show consistent improvements across image classification, event-based recognition, and semantic segmentation. In particular, QKFormer with SLI and ACF achieves 84.37% Top-1 accuracy on ImageNet-1K and 37.5% mIoU on ADE20K, where the segmentation model is trained without ImageNet pretraining. Ablation studies and qualitative analyses further indicate that SSA and SLI capture complementary interaction patterns and that learnable fusion consistently outperforms fixed weighting.
[CV-74] Flow Matching-Based PET Image Reconstruction
链接: https://arxiv.org/abs/2608.20112
作者: Fumio Hashimoto,Ziqian Huang,Tatsuya Yokota,Kuang Gong
类目: Image and Video Processing (eess.IV); Computer Vision and Pattern Recognition (cs.CV); Medical Physics (physics.med-ph)
备注: 10 pages, 8 figures
Abstract:Generative models have shown strong potential for positron emission tomography (PET) image reconstruction. Although diffusion model-based reconstruction methods have demonstrated promising performance, they often require many reverse sampling steps with data-consistency updates incorporated into the sampling process. Flow matching offers an attractive alternative because it can directly estimate clean images from intermediate states, allowing data-consistency refinement to be separated from flow propagation. In this work, we proposed flow matching-based PET image reconstruction methods. We first established PET-FlowDPS by incorporating Poisson likelihood guidance with an expectation-maximization (EM)-based preconditioner into the FlowDPS framework. We then proposed a model-based PET reconstruction method that used a pretrained flow matching model as a prior, in which the flow-based prior, PET data refinement, and stochastic propagation were interpreted within an approximate Bayesian framework. Experimental results using [ ^\text18\textF ]FDG brain PET datasets showed that the proposed method achieved better bias-variance trade-offs across different dose levels compared with other reference methods. These results demonstrated the potential of flow matching as a generative prior for quantitative PET image reconstruction.
[CV-75] MOSAIC: Modality-agnostic Spectral Alignment for Federated Image-level Weakly Supervised Tumor Segmentation under Client-specific Missing Modalities
链接: https://arxiv.org/abs/2608.19788
作者: Tarun Kumar Garg,Vaanathi Sundaresan
类目: Image and Video Processing (eess.IV); Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Trustworthy multimodal fusion in clinical settings requires handling incomplete and heterogeneous modality subsets across institutions, where privacy constraints prohibit centralized data sharing. Federated learning (FL) mitigates data-sharing constraints but suffers from client-specific missing modalities, where institutions possess incomplete multimodal subsets, degrading fusion quality and segmentation performance. While FL and weak supervision have been studied separately, their joint use with image-level labels under heterogeneous missing modalities remains unaddressed. We propose \textbfMOSAIC, the first modality-agnostic federated framework for weakly supervised binary tumor segmentation under client-specific missing modalities. We introduce a client-specific modality-alignment module that fuses available channels into a shared latent space without prior knowledge of modality identity, a spectral prototype alignment loss that reconciles cross-client distribution shift using compact non-invertible frequency-domain statistics, and a dedicated federated refinement network that denoises the resulting CAM pseudo-labels into accurate masks, breaking the accuracy ceiling of weak supervision. Experiments on three multi-institutional brain tumor benchmarks (FeTS2022, BraTS-MEN, and BraTS-SSA) demonstrate significant improvements over all image, box, and point-supervised baselines, approaching fully supervised accuracy using only image-level labels and reaching 0.84 Dice on FeTS2022. Dynamic new client addition enables previously unseen institutions to join an already-trained federation within 0.01-0.04 Dice without retraining. Code is available at this https URL.
[CV-76] AsymFeX: A Symmetry-Driven Framework for Ischemic Stroke Segmentation Across Imaging Modalities and Stroke Stages
链接: https://arxiv.org/abs/2608.19769
作者: Maunil Shah,Vaanathi Sundaresan
类目: Image and Video Processing (eess.IV); Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Fast and accurate segmentation of Acute Ischemic Stroke (AIS) lesions is essential for stroke prognosis and treatment planning. Non-contrast CT (NCCT), the first-line imaging modality for diagnosing ischemic infarcts, exhibits subtle infarct contrast, making manual delineation slow and labor-intensive. Motivated by this, and by the clinical practice of comparing brain hemispheres to localize infarcts, we propose a two-stage, nnU-Net-compatible 3D segmentation method. The first stage corrects head tilt to align each scan to its true anatomical mid-sagittal plane; the second applies a novel Asymmetric Feature Extraction (AsymFeX) module, comparing each voxel to its true contralateral counterpart within a local 3 x 3 x 3 neighborhood via cross-hemispheric attention, feature disparity estimation, and dual-scale gating to capture both large and small infarcts. On AISD, our method achieves 0.6796 Dice, 23.53 mm HD95, and 7.69 mL AVD, significantly outperforming existing state-of-the-art methods, with clinically relevant volumetric analysis at the 70 mL thrombolysis-eligibility threshold. Proof-of-concept evaluation on ATLAS v2.1 and ISLES’24 demonstrates that the same symmetry-driven design generalizes across imaging modalities and stroke time points without architectural changes, further supported by an uncertainty analysis assessing reliability under clinical deployment. Code is publicly available at this https URL.
人工智能
[AI-0] Pandoras AI Model Routing Box: Efficient Allocation with Costly Value Estimation
链接: https://arxiv.org/abs/2608.20316
作者: Adam Fisch,Shubhendu Trivedi,Fantine Huot,William W. Cohen,Michael Kaisers,Mirella Lapata,Kate Larson,Jacob Eisenstein
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Heterogeneous AI systems composed of multiple models, architectures, harnesses, or inference-time settings can improve quality and efficiency by routing queries to the specialist who can answer most effectively at the lowest cost. Routing requires estimating each specialist’s expected return, but this value estimation has a cost. Cheap estimators (e.g., embedding-based predictors) are fast but noisy, while accurate estimators (e.g., fine-tuned models with access to retrieval results or partial reasoning traces) are expensive. We formalize this tradeoff as an instance of Pandora’s Box, the classical problem of optimal search with costly inspection. Under a Gaussian signal model, the resulting policies have closed-form value-of-information expressions that determine, for each specialist and input, whether refining the value estimate is worth its cost. We call the centralized policy Pandora’s Router. We extend this to a decentralized setting, Pandora’s Bidder, where specialists independently decide whether to invest in self-assessment before accepting an offered price to claim a query. Experiments across three domains—a standard multi-LLM benchmark, retrieval-augmented specialists, and LLMs with variable inference-time reasoning—show that Pandora’s Router matches the routing quality of exhaustive estimation, while querying the expensive estimator far less often. In the decentralized setting, value-of-information reasoning improves allocative efficiency when competing estimates are accurate; when competing estimates are noisy, however, it can increase the strategic specialist’s utility at the expense of others.
[AI-1] MidTool: Mid-training Data Synthesis for Agent ic Tool Use
链接: https://arxiv.org/abs/2608.20314
作者: Fengqing Jiang,Yite Wang,Boyi Liu,Zhaoyang Wang,Canwen Xu,Zhewei Yao,Radha Poovendran,Yuxiong He
类目: Artificial Intelligence (cs.AI)
备注: Data Model: this https URL
Abstract:Mid-training is increasingly recognized as a critical stage for shaping the capabilities of large language models. Recent work has shown that targeted mid-training can strengthen reasoning-intensive abilities such as math and science, and can also improve agentic capabilities in software-engineering settings. In this work, we study the parallel but less explored agentic capability: general tool use. We present MidTool, an open corpus construction pipeline for agentic tool-use mid-training that combines large-scale web, PDF, and code data with synthesized supervision from real-world tool APIs, MCP skills, and document-grounded workflows. MidTool is designed to teach models how to recognize tool affordances, ground arguments from context, compose tool call workflow, and recover from incomplete information. We mid-train Qwen3-4B-Base and Qwen3-8B-Base on MidTool-Mix, and then apply follow-up post-training with both supervised fine-tuning and reinforcement learning. Compared with baselines, MidTool-Mix consistently improves downstream performance under both SFT and RL on BFCL, tau2-Bench, and MCP Universe. These results suggest that general tool use, like other important LLM capabilities, benefits from dedicated mid-training rather than being left entirely to post-training.
[AI-2] Catching the Rug: Early Prediction of Fraudulent Memecoins on Solana via Machine Learning
链接: https://arxiv.org/abs/2608.20271
作者: Jianghai Li,Pavel Kuznetsov,Yury Yanovich,Konstantin Nott-Whaley,Igor Vodolazov
类目: Artificial Intelligence (cs.AI); Distributed, Parallel, and Cluster Computing (cs.DC)
备注:
Abstract:The rapid proliferation of memecoins on blockchain platforms has increased the risk of fraudulent activities, particularly rug pulls. While previous studies have focused on Ethereum-based tokens, this paper shifts the spotlight to Solana, the leading blockchain for memecoins by trading volume and token count. Unlike Ethereum, where rug pulls often exploit smart contract backdoors, Solana memecoin rug pulls are predominantly driven by liquidity manipulation and social dynamics. This research pioneers large-scale rug pull early detection in the Solana ecosystem by assembling a dataset of 6.4 million tokens over 7 months. Market analysis reveals that a vast majority of these memecoins exhibit rug pull characteristics within one hour of launch, highlighting the urgency of short-horizon prediction. Despite the absence of code-level features, we demonstrate that classic machine learning models, particularly Gradient Boosting (XGBoost), achieve robust performance in detecting potential rug pulls using only the first 5 minutes of trading data. Furthermore, we evaluate cross-platform generalization between PumpFun and Raydium, revealing that multi-source data fusion significantly mitigates domain shift and improves detection reliability. This study advances the understanding of DeFi fraud on high-throughput chains and provides a practical framework for protecting investors.
[AI-3] Learning When to Think: Adaptive Reasoning for Test-Time Compute Allocation
链接: https://arxiv.org/abs/2608.20256
作者: Gijs Kassenaar,Zhao Yang,Vincent François-Lavet
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Reasoning language models trained with reinforcement learning typically operate under a fixed token budget rather than an explicitly adaptive one, which can lead to over-computation on easy problems and insufficient computation on difficult ones. We study whether a model can learn to allocate its own reasoning effort by choosing, as the first token of its response, one of three modes: \textscNoThink (answer as quickly as possible), \textscShort (brief reasoning), or \textscLong (extended reasoning). The choice is learned inside Group Relative Policy Optimization (GRPO) with no separate router, through a shaped reward that makes each mode worthwhile at a different response length, together with hard per-mode token caps that keep the modes distinct. On a 1.5B distilled model trained on MATH, the three modes emerge without collapsing to a single choice, and the brief modes end up more accurate than \textscLong, which shows that the router sorts problems by difficulty rather than at random. Averaged over three seeds, the resulting policy stays close to the base model’s accuracy on the held-out MATH500 ( 0.782 vs.\ 0.796 ) while cutting the mean response length from 4,796 to 2,811 tokens (a 41% reduction). Interestingly, it also transfers to other benchmarks without retraining, with the largest savings where problems are easier, with for instance 76% token reduction on GSM8K and at higher accuracy than the baselines at similar response length. In short, we build a reasoning model that adaptively chooses how much to reason for each problem.
[AI-4] QUASAR: A Quantum-Classical Neural Network for SAR Satellite Physical-Layer Authentication NDSS2027
链接: https://arxiv.org/abs/2608.20240
作者: Vincenzo Sammartino,Nathanael Denis,Roberto Di Pietro
类目: Artificial Intelligence (cs.AI); Cryptography and Security (cs.CR)
备注: Submitted to NDSS 2027
Abstract:X-band SAR satellites (8-12 GHz) play a critical role in disaster response, environmental monitoring, and military intelligence. Yet, they lack robust physical-layer authentication (PLA), a security layer orthogonal to cryptographic solutions. Existing PLA systems, typically based on radio-frequency fingerprinting, are often limited to sub-6 GHz frequencies and rely on classical deep learning. However, this approach underfits the IQ phase nonlinearities that distinguish satellite hardware. In this paper, we present QUASAR, to the best of our knowledge the first quantum-classical hybrid architecture that fuses a CNN spectrogram encoder with a variational quantum circuit (VQC) to provide PLA to X-band SAR signals. Our solution enjoys two distinctive features: (i) it is markedly more data-efficient than classical machine learning, requiring only 10% of the training data to match the accuracy of classical baselines – data collection being notoriously the most time-consuming phase of PLA; and, (ii) at an equal data budget, it improves classification accuracy over those baselines. In detail, we test our solution under three adversarial scenarios: replay, crafted-IQ injection, and space-borne spoofing. QUASAR rejects spoofed transmissions in 89.7%, 94.1%, and 81.3% of attempts, respectively, establishing the first quantum-enhanced physical-layer classifier for satellite constellations. The fully detailed framework and the supporting results, other than being interesting on their own, show a novel research avenue for physical-layer authentication.
[AI-5] Rule-Compliant Visual Spatial Planning for Multimodal Large Language Models
链接: https://arxiv.org/abs/2608.20237
作者: Yu Chen,Ting Lei,Yaoyi Li,Jia Cai,Zhecen Wu,Yang Liu
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Multimodal large language models (MLLMs) combine linguistic reasoning with visual perception, yet their ability to perform visual spatial planning under explicit or previously unseen rule constraints remains underexplored. This setting requires models to jointly understand spatial layouts, interpret natural-language rules, and plan valid actions accordingly. To address this gap, we introduce RuleMaze, a controllable benchmark in which MLLMs must navigate mazes while obeying natural-language rules of varying complexity. RuleMaze isolates rule-compliant spatial planning by requiring accurate perception, rule interpretation, and constrained action planning. To enable scalable and systematic rule construction, we propose Language-Logic-Function Hybridization, which automatically generates natural-language rules and translates them into logical representations and executable validators, eliminating manual rule engineering. To improve rule following and generalization, we introduce Disentangled Multimodal Planning (DMP), which separates perception, execution, and rule verification through interpretable reasoning primitives. By disentangling these components, DMP facilitates systematic generalization to more complex and previously unseen rules, while providing transparent intermediate planning traces. Experiments demonstrate that DMP substantially improves rule compliance and planning success compared to end-to-end textual planning baselines. Overall, RuleMaze establishes a principled benchmark for studying grounded and interpretable rule-based spatial planning in MLLMs. Code is available at this https URL.
[AI-6] InsufficiencyBench: Evaluating LLM legal advice on underspecified user queries ICML
链接: https://arxiv.org/abs/2608.20220
作者: Samuel J. Vincent,Daniel Calloway,Fangyi Yu,Andrew M. Bean,Nabeel Seedat
类目: Artificial Intelligence (cs.AI)
备注: 10 pages, Best Paper Honorable Mention at ICML AI4Law 2026
Abstract:Legal AI systems are increasingly used to answer legal questions, yet existing benchmarks assume queries arrive fully specified. In practice, users omit facts that materially determine the legal outcome. We introduce InsufficiencyBench, the first legal benchmark targeting query-side insufficiency: whether a model recognizes when a query lacks legally material information, identifies what is missing, and refrains from premature conclusions. We formalize a taxonomy of eight canonical missing-element categories across three structural failure modes—switch, gating, and fatal prerequisite— and construct 202 benchmark items (58 base queries, 144 deficient variants) spanning six legal domains and 24 US jurisdictions and annotated by practising attorneys. Evaluating ten frontier models, we find that no model exceeds F2 = 0.46 on missing-element identification and that the median recall is 0.44. Models either hedge indiscriminately or answer silently under fabricated presumptions. No model both identifies and qualifies responses to deficient queries while directly addressing complete ones.
[AI-7] Electronic Navigational Chart Change Classification
链接: https://arxiv.org/abs/2608.20218
作者: Jacob Arndt,Abhishek Potnis,Alexandre Sorokine
类目: Artificial Intelligence (cs.AI)
备注: Accepted at 34th ACM SIGSPATIAL 2026
Abstract:Electronic Navigational Charts (ENCs) are geospatial vector datasets used in maritime navigation systems that represent hydrographic and navigational information such as depths, navigational aids, traffic schemes, and hazards. A major challenge for hydrographic offices is determining whether a given chart change poses a critical or non-critical risk to maritime safety. Existing workflows rely heavily on manual review and verification, which is labor-intensive, scales poorly with the volume of incoming chart updates, and introduces inter-analyst inconsistencies. To address this challenge, we propose a method for automated classification of ENC changes. We establish a baseline encoding scheme to translate complex vector data changes into a structured tabular format for classification models. The two crucial components of the encoding scheme include a spatial context encoder to enrich the change representations with surrounding geographic features, and an ENC attribute encoder to represent nuanced attribute-value descriptions of the modified objects. We evaluate the proposed approach across two distinct operational datasets, comprising 1,308 chart pairs containing over 100,000 individual chart modifications. Tuned gradient-boosted trees leveraging the proposed encoding schemes achieve accuracies of 90% and 94% on the two datasets, yielding a 5-7% improvement over default hyperparameterized models trained on encodings without spatial context and attribute embeddings. These results demonstrate the viability of integrating machine learning into operational geospatial pipelines to improve ENC maintenance and enhance maritime safety. Finally, our experiments demonstrate the effectiveness of simple location and spatial aggregation methods, providing a foundation for evaluating more sophisticated spatial representation learning techniques for this application.
[AI-8] he Third Restructuring of Software Form: From the Three-Tier Architecture to Storag e Models and Agents
链接: https://arxiv.org/abs/2608.20201
作者: Wei Lin,Tao Zhou,Zhaofei Xie,Changgui Hong
类目: Artificial Intelligence (cs.AI); Software Engineering (cs.SE)
备注:
Abstract:Software form has undergone two paradigm shifts since its inception: Software 1.0, in which instructions determine behavior, and Software 2.0, in which data determines behavior (machine learning). This paper argues that a third shift - Software 3.0, in which context and reasoning determine behavior - is now underway, and contends that its terminal form converges to three elements: a generalized database (the unified abstraction of all persistent state and memory), a large model (the intelligence core that performs reasoning and generation), and an agent (the execution loop connecting the first two). The core argument is as follows: in the traditional three-tier architecture, the user-interface layer will be absorbed by the model’s ability to generate interfaces on demand, the business-logic layer will be re-partitioned along “expressibility x criticality” into model reasoning and storage constraints (with residual deterministic logic retained as tools), and only the data layer will be elevated into the sole persistent infrastructure. We formalize this convergence thesis, present a minimal reference architecture, report evidence from real prototypes and a live model, and systematically analyze both the conditions under which it holds and the boundaries where it fails - determinism, cost, security, and verifiability delimit the thesis’s domain of applicability. We argue that the thesis holds in task domains that are expressible, verifiable, externally stateful, and tool-complete, and that it will reshape the roles of developers, the database industry, and the software-engineering discipline.
[AI-9] A Standardized Framework for Machine Learning in Power System Protection
链接: https://arxiv.org/abs/2608.20181
作者: Julian Oelhaf,Georg Kordowich,Paula Andrea Pérez-Toro,Christian Bergler,Johann Jäger,Andreas Maier,Siming Bayer
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Signal Processing (eess.SP)
备注: 32 pages, 4 figures, 26 tables. Code: this https URL . Dataset: PROTECT-90, doi: https://doi.org/10.5281/zenodo.21109169
Abstract:Studies of machine-learning-based power-system protection increasingly report near-perfect scores, yet the meaning of those scores depends strongly on the evaluation setting. Protection task, physical scope, measurements, timing, targets, preprocessing, and validation often vary jointly and remain incompletely specified. This paper proposes a standardization-oriented framework that treats evaluation design as part of the scientific contribution. It defines seven required study dimensions: protection objective, physical scope, observability, timing and decision windows, targets and sample validity, validation protocol, and evaluation outputs. The framework is instantiated in a bounded case study on the public PROTECT-90 electromagnetic-transient benchmark, comprising 9022 simulated episodes from a 90 kV double-line topology, for onset-conditioned fault classification and localization. Under centralized sensing, simulation-metadata-aligned 20 ms windows, and episode-grouped validation, a multi-layer perceptron (MLP) achieved a five-fold mean macro-averaged F1 score of 0.991 +/- 0.001 for classification and a localization mean absolute error of 10.20 +/- 0.25% of line length (mean +/- std across episode-grouped folds). Extending the decision horizon to 50 ms preserved this task-dependent performance asymmetry, while reduced observability approximately doubled the MLP localization error but had little effect on classification. A synchronized two-ended conventional locator outperformed the learning locators under its richer clean information set, and measurement degradation showed that clean predictive performance did not determine robustness. The framework turns evaluation assumptions into explicit, reproducible evidence and provides a basis for more comparable, auditable evaluation and future certification-oriented assessment of machine-learning protection functions.
[AI-10] DARS: Dual-Level Credit Assignment RL with Structured Reasoning for Instruction-Based Image Editing
链接: https://arxiv.org/abs/2608.20161
作者: Haoxiang Cao,Jiajiong Cao,Xuanpu Zhang,Changqian Yu,Chaoqun Wang
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Instruction-based image editing uses a planner-renderer pipeline: a vision-language model (VLM) first converts the instruction into an edit plan, and a diffusion model then executes that plan. Training such systems with only final-image rewards is inefficient because a poor edit does not reveal whether additional optimization should place more emphasis on the planner or the renderer, and even planner-dominant cases remain difficult to localize within a free-form reasoning trace. We present DARS, a reinforcement learning framework for dual-level credit assignment in this two-stage setting. Across modules, multi-plan multi-render rollouts estimate between-plan and within-plan reward variability for soft module routing, while rollout mean rewards provide hardness estimates for an adaptive curriculum. Within the planner, a four-field structured reasoning output enables a prefix-gated reward and token-level advantage reweighting, turning outcome-level feedback into localized supervision. Experiments on five benchmarks show that DARS outperforms a Joint~RL baseline with the same backbone, data, reward model, and rollout budget, with the largest gains on reasoning-intensive edits.
[AI-11] DECOWAM: Decoupled Whole-Body World-Action Model for Legged Mobile Manipulation
链接: https://arxiv.org/abs/2608.20114
作者: Siyuan Ma,Boshi Zhang,Yutian Zhang,Qinglian Wu,Jiaqi Zhai,Dong Wei,Qiaojun Yu
类目: Artificial Intelligence (cs.AI); Robotics (cs.RO)
备注: 8 pages, 5 figures. Introduces DECOWAM, a decoupled whole-body world-action model for legged mobile manipulation, and the ARMDOG real-robot dataset
Abstract:Mobile manipulation requires a robot to predict how locomotion and arm motion jointly alter future observations and control. Existing world-action models, developed largely for fixed-base platforms, do not explicitly distinguish camera ego-motion from base and arm actions. Here we introduce DECOWAM, a whole-body world-action model that separates these factors through dedicated conditional interfaces. DECOWAM freezes an adapted FastWAM backbone and trains residual adapters, an action-equivalent future bottleneck distilled from privileged observations, adversarially separated base and arm latents, and base-velocity conditioning for video prediction. We further introduce ARMDOG, a real-robot dataset that synchronizes video, whole-body state and action, and language. On a fixed replay protocol, DECOWAM improved both future-video and action prediction over FastWAM, reducing action MSE by 21.7% with 25.95M trainable adaptation parameters. Across 79 closed-loop trials per method, it achieved the highest observed whole-body coordination and base-displacement robustness among the compared systems, while task completion remained comparable to the strongest baseline. These results show that embodiment-aware factorization can support parameter-efficient joint visual prediction and whole-body control under moving viewpoints.
[AI-12] owards Professional Tennis Styles for Humanoid Robots with Adaptive Motion Planning and Tracking
链接: https://arxiv.org/abs/2608.20087
作者: Tao Huang,Ruofei Liu,Xuchen Tang,Xinyin Zhang,Junli Ren,Huayi Wang,Feiyu Jia,Yukai Qi,Kangning Yin,Weishuai Zeng,Lipeng Chen,Xi Li,Ting Wu,Kailin Li,Ruoli Dai,Jingbo Wang,Lei Han,Jiangmiao Pang
类目: Robotics (cs.RO); Artificial Intelligence (cs.AI)
备注: 14 pages
Abstract:Humanoid robots have recently demonstrated promising capabilities in real-world ball sports. However, achieving professional motion styles while maintaining strong task performance remains challenging. In this work, we propose AdaPT, an Adaptive Motion Planning and Tracking framework that learns professional tennis serving and rally styles directly from broadcast videos. This hierarchical design is motivated by the key insight that the planner generates stylistic kinematic motions, while the tracker executes them with minimal interference with planning. Despite its effectiveness in simulation, a substantial sim-to-real gap emerges: tracking performance inevitably degrades on real robots, and this degradation is partially overlooked by autoregressive planning and further compounded by noisy perception. To address these issues, our adaptation mechanism improves tracking robustness by learning to track randomized execution speeds, while conditioning the planner on a learned motion-speed adapter to mitigate compounding errors. Real-world experiments on the Unitree G1 demonstrate the effectiveness of our adaptation mechanism in bridging the sim-to-real gap. We further deploy AdaPT policies on the full-size Dobot Atom humanoid robot (1.7m) and demonstrate in-the-wild serving without motion capture. Beyond these results, our real-world experiments reveal both algorithmic and engineering insights for future humanoid ball-sports systems. Videos and code are available on our \hrefthis https URLproject website.
[AI-13] Evidence-Gated Task and Motion Planning with Vision-Language Models
链接: https://arxiv.org/abs/2608.20084
作者: Tsunehiko Tanaka,Matthew Stephenson,Alistair Macvicar,Edgar Simo-Serra
类目: Robotics (cs.RO); Artificial Intelligence (cs.AI)
备注:
Abstract:Robots executing long-horizon manipulation tasks from natural-language instructions must reason about both semantic task structure and geometric feasibility. However, under partial observability, the availability of goal-relevant objects may be uncertain. In such cases, approaches that combine Vision-Language Models (VLMs) with Task and Motion Planning (TAMP) may generate subgoals that rely on the VLM’s prior knowledge without observational support, leading to execution failures or unintended outcomes. We propose Evidence Acquisition and Feasibility Gating (EAFG), a framework that acquires visual evidence through VLM-generated exploratory subgoals and TAMP-based execution. EAFG then applies a feasibility gate to decide whether to proceed with task planning, acquire further evidence, or halt. Our experiments show that, in cooking tasks with ambiguous object use, EAFG improves recipe completion by discovering task-relevant objects before planning. For instructions requiring an absent object, EAFG promotes appropriate halt decisions and reduces repeated attempts to manipulate that object.
[AI-14] EchoCoT: Extracting Hidden Chain-of-Thought from Large Reasoning Models
链接: https://arxiv.org/abs/2608.20055
作者: Yiting Qu,Ziqing Yang,Chi Cui,Ye Leng,Junjie Chu,Yang Zhang
类目: Cryptography and Security (cs.CR); Artificial Intelligence (cs.AI)
备注:
Abstract:Hidden chain-of-thought (CoT) traces, especially those from frontier proprietary large reasoning models (LRMs), are valuable model assets. Yet whether these hidden CoTs can be directly extracted from black-box models remains largely unexplored. In this work, we systematically study whether hidden CoTs can be extracted near-verbatim from black-box LRMs through API interactions. We identify a previously overlooked reasoning replay surface between tool calls and develop EchoCoT, a multi-step attack that iteratively extracts hidden CoTs using API-returned fidelity signals. We further develop an LLM-based optimization framework that automatically searches for an effective universal injection trajectory across various datasets. We evaluate EchoCoT on three open-source and five frontier proprietary LRMs. On open-source LRMs, EchoCoT achieves up to 66.4% near-verbatim extraction success, with the extracted trace length within 10% of the target and at least 90% of tokens exactly matching the target CoT. The same injection trajectory also generalizes to unseen datasets, achieving up to 80% extraction success under the same criterion. For tested frontier proprietary LRMs, a substantial fraction of extracted CoTs closely align with provider-reported reasoning lengths and available CoT summaries. EchoCoT can also extract very long CoTs: on Gemini-2.5, it extracts 33,463 tokens from a 32,948-token target. These results establish hidden-CoT extraction as a practical security risk and highlight the need to better protect hidden CoT assets.
[AI-15] On the Applicability of Safety Nets: A Safety-By-Design Solution for Certifying Neural Networks
链接: https://arxiv.org/abs/2608.20053
作者: Johann Maximilian Christensen,Thomas Stefani,Elena Hoemann,Frank Köster,Sven Hallerbach
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:The integration of Artificial Intelligence (AI) in safety-critical aviation systems presents significant challenges for certification and deployment. Aviation, often regarded as the safest form of transportation, relies on numerous safety-critical systems. For future safety-critical AI-based systems, EASA requires a Safety-by-Design approach, which can be achieved by using Safety Nets that combine neural network compression with lookup tables to ensure 100 % correct runtime behavior across the discretized operational design domain. Although Safety Nets have been studied, no comprehensive study of their performance characteristics and system design trade-offs has been conducted. This work presents the first systematic analysis of the trade-off between neural network and lookup table size in Safety Nets. By systematically comparing neural networks with diverse architectures, this study identifies optimal design parameters that minimize overall storage and memory requirements while maintaining certification compliance. Results demonstrate that architectures with 3 to 5 hidden layers, each with approximately 50 to 100 nodes, combined with one-hot encoding, achieve the best balance. In these configurations, neural networks accurately represent at least 97 % of the data, while compact lookup tables handle the remaining errors. The resulting Safety Nets reduce the system size by almost three orders of magnitude, fitting within the memory budget of current avionics hardware while guaranteeing 100 % correct outputs across the entire discretized input space, as required by EASA guidelines. This work provides the first-ever open-source implementation of Safety Nets for HCAS and VCAS with replicable results, demonstrating a practical pathway toward certifiable AI-based systems in aviation and establishing Safety Nets as a viable Safety-by-Design solution for safety-critical applications.
[AI-16] A three-dimensional typology of agency for advanced AI systems
链接: https://arxiv.org/abs/2608.20041
作者: Willem Fourie
类目: Artificial Intelligence (cs.AI); Computers and Society (cs.CY)
备注:
Abstract:Research on the agency of advanced artificial intelligence (AI) systems focuses on agency as a normative concept and on the agency of particularly agentic AI systems. While recent work also focuses on the different profiles of agentic systems, no framework exists to address the question of the type of agency instantiated by advanced AI systems, particularly when considering non-moral forms of agency. Based on established theoretical positions in philosophy, ethics, legal theory and sociology, we develop a typology of agency for frontier AI systems consisting of three dimensions: the nature of agency (moral or legal), its mode (individual or collective) and its locus (human or non-human). Combining these dimensions produces eight possible instantiations of agency, which we classify as conventional, contested or controversial. The typology separates legal from moral agency and thereby creates conceptual space for considering individual, legal, non-human agency without presupposing that advanced AI systems are moral agents. We argue that this distinction is increasingly relevant where instrumental goal pursuit complicates the attribution of AI actions to particular human actors.
[AI-17] Contrastive Mixed Prompt Learning for Incomplete Multimodal Sentiment Analysis with Unseen Modality Combination
链接: https://arxiv.org/abs/2608.20019
作者: Kaixin Xu,NaiJin Liu,Yulin Kang,Tangyue Jin,Zixuan Yu,Wenxi Zhao,Yibei Liu,Qianle Zhang,Yangyang Wu,Mengying Zhu,Meng Xi
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Incomplete multimodal sentiment analysis has garnered significant attention in recent years. Existing approaches typically assume that data is missing at random or are designed specifically for certain missing patterns, ignoring the modality combination inconsistency between training and testing phases. However, in real-world scenarios, the testing phase often encounters modal combinations that were not present during the training phase, which leads to insufficient generalization capabilities and unstable performance. In this paper, we introduce the problem of Incomplete Multimodal Sentiment Analysis with Unseen Modality Combinations (IMSAUMC), aiming to enhance model generalization for unseen modality combinations. To address this challenge, we propose the model named \textbfC ontrastive \textbfM ixed \textbfP rompt \textbfL earning ( \textsfCMPL ) for IMSAUMC. It introduces a label-guided contrastive feature learning mechanism to learn robust and discriminative cross-modal representations. Additionally, we design modality-combination prompts with a soft router to facilitate better learning of various modality combinations. Furthermore, we introduce three prompt contrastive learning strategies, which enable effective learning of prompts corresponding to unseen modality combinations, thereby significantly strengthening the model’s generalization capabilities in diverse testing scenarios. Extensive experiments on three widely used datasets demonstrate that \textsfCMPL achieves more than a 5% improvement in accuracy compared to state-of-the-art approaches.
[AI-18] ExPhy: A Benchmark for Explicit Physical Property Learning in Multi-Object Trajectory Forecasting
链接: https://arxiv.org/abs/2608.20009
作者: Rui Wang,Yeteng Wu,Xianlin Zhang,Mengshi Qi
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Understanding object dynamics requires not only predicting future trajectories but also examining whether a model captures the physical properties that govern motion. However, existing benchmarks rarely expose object-level physical properties as explicit evaluation targets alongside trajectory forecasting. To address this gap, we introduce \emphExPhy, a multi-object trajectory forecasting benchmark containing 24,000 simulated physical scenes with explicit object-level labels for mass, friction, and restitution. ExPhy provides observed and future trajectories together with an in-distribution (ID) split and two out-of-distribution (OOD) splits over physical parameters (OOD-Parameter) and initial states (OOD-Initial) for jointly evaluating trajectory forecasting and physical property estimation. We further instantiate \textscPhyODE, a physics-guided model with an explicit property interface that estimates physical properties from observed trajectories and uses them for differentiable future rollout. On the long-horizon OOD-Initial setting, \textscPhyODE reduces ADE and FDE by 33.1% and 31.0%, respectively, compared with the strongest baseline. Zero-shot evaluation on ComPhy further assesses cross-benchmark transfer. Property-level analyses reveal that accurate trajectory forecasting does not necessarily imply accurate recovery of the underlying physical properties. Code and data are available at this https URL.
[AI-19] Optimal Skill Selection for LLM Agents with Provable Bicriteria Guarantees
链接: https://arxiv.org/abs/2608.19993
作者: Yu Chen,Ruishuo Chen,Xun Wang,Zhuoran Li,Longbo Huang
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Loading reusable skill documents into a bounded context window is now the primary way large language model (LLM) agents acquire task-specific capabilities, which makes skill selection a first-order determinant of task performance and token cost. Yet current agents score skills independently by semantic relevance and assemble the set by top- k or greedy packing, with no quality guarantee or cost awareness on the selected set. As a result, redundant or poorly chosen skills waste scarce context tokens and can even degrade performance. We give the first model of how the selected skill set shapes execution outcomes and cast skill selection as an optimization problem: choose a skill set under a hard token budget to maximize a monotone submodular benefit minus context penalty. For this problem, we develop Best Prefix Selection (BPS), a polynomial-time algorithm, and prove, to our knowledge, the first performance guarantee for skill selection: a bicriteria (1-1/e,1) approximation whose benefit coefficient is optimal in polynomial time. On a contamination-controlled BigCodeBench variant, BPS outperforms all the baselines, reaching 0.73 measured task success versus 0.20 – 0.52 for released skill routers, text retrievers, and the executor’s own selection, on 28% fewer tokens than the strongest released router.
[AI-20] ReguSim: Evaluating LLM Agent Rule Grounding in Financial Compliance
链接: https://arxiv.org/abs/2608.19974
作者: Yiyang Luo,Yihang Jiang,Qijun Xie,Liang Lan,Lin Willian Cong,Anyi Rao,Yunya Song
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:LLM agents in financial markets may cite rules yet still submit orders that violate executable constraints or misread surveillance evidence. We introduce ReguSim, a controlled financial-compliance environment, and ReguBench, a target-marked monitoring benchmark, to separate four artifacts: stated reasoning, attempted action, execution enforcement, and monitor evidence. In trader runs with DeepSeek V4 Pro and Gemini 3.5 Flash, visible rules reduce but do not eliminate rejected actions, and incentive or persona framing shifts behavior. A bridge study shows that trader rationales can mislead an independent monitor unless enforcement evidence is shown. In monitoring, simple structured baselines either match or exceed prompt-only LLMs. The results frame financial compliance evaluation as an audit of rule-grounded actions and evidence use, rather than a single compliance score.
[AI-21] Rethinking Patch Based Multivariate Time Series Forecasting with Semantic Structured Partitioning
链接: https://arxiv.org/abs/2608.19966
作者: Jiazhe Wang,Zhiquan Huang,Linjing Xue,Ming Liu,Meiwen Li,Ruijuan Zheng
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Multivariate time series forecasting (MTSF) is a fundamental task in many real world applications. Existing patch based forecasting methods generally fall into three categories: fixed partitioning, multi-scale partitioning, and extendable partitioning. Fixed partitioning often breaks meaningful temporal boundaries, multi-scale partitioning may introduce redundant representations across scales, and extendable partitioning improves flexibility but still lacks an explicit mechanism for organizing semantic structure and modeling interactions among heterogeneous temporal patterns. To address these limitations, we propose SCPaT, a Transformer based framework built on semantic structured partitioning. SCPaT first decomposes input sequences into semantically consistent units through adaptive semantic unit generation, then constructs a dynamic semantic graph to model directed dependencies among these units and organize them into higher order semantic blocks. Based on these structured representations, an importance aware routing mechanism adaptively dispatches different semantic blocks to different experts for customized modeling. Extensive experiments on 12 real world datasets demonstrate the effectiveness of SCPaT.
[AI-22] Learning Early-to-Final Solution Consistency for MILP Acceleration
链接: https://arxiv.org/abs/2608.19953
作者: Guanlin Li,Chengrui Gao,Chenguang Wang,Haopu Shang,Zherong Zhang,Ke Xue,Jixiang Lu,Weiyong Yang,Chao Qian
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Mixed-Integer Linear Programming (MILP) is a fundamental problem class in operations research and combinatorial optimization, with broad applications to industrial decision-making. Owing to their NP-hardness, however, modern solvers may struggle to find high-quality solutions for challenging MILP instances within practical time limits. Recent learning-based approaches seek to accelerate MILP solving by directly predicting high-quality solutions from static instance-level features, such as variable-constraint bipartite graphs. Yet accurate solution prediction from instance features alone is difficult, and these methods largely overlook the information revealed during the solver’s search process. In this paper, we find that solutions produced at the early search stage of MILP solvers, which are computationally cheap to obtain, are often structurally close to the solutions found after full-budget search. Motivated by this observation, we propose a new solver-informed paradigm that shifts the learning target from variable assignment to early-to-final consistency: for each variable, we predict whether its early-stage assignment should persist in full-budget solutions. The predicted consistency naturally guides downstream search, for instance by fixing the assignments deemed consistent. At inference time, we further ensemble consistency predictions across multiple early-stage solutions to improve robustness. Experiments across four MILP benchmarks show our method improves prediction-guided search across diverse downstream pipelines. With Gurobi, our proposed method reduces the primal gap by 56.9% on average and closes it completely on combinatorial auction instances. Besides, we transferred the Gurobi-trained model zero-shot to SCIP without adaptation, achieving a 36.4% average gap reduction across benchmarks.
[AI-23] owards Quantifying Benchmark Optimization in ASR Models
链接: https://arxiv.org/abs/2608.19936
作者: Theo Lebryk,David Ayllon,Alice Baird,Jakub Piotr Cłapa,Jens Madsen,Panagiotis Tzirakis
类目: ound (cs.SD); Artificial Intelligence (cs.AI)
备注:
Abstract:Public benchmarks are important measures of Automatic Speech Recognition (ASR) model capabilities. However, by nature of being public, there is risk of models being optimized for these benchmarks in ways that do not generalize well to real-world data. We present a methodology for quantifying benchmark optimization, focusing on cases where the audio underdetermines the reference transcript. We identify three families of behavioral probes that reveal models’ capabilities of reproducing benchmark reference spans despite underdetermined audio: reference disagreement, masked-number recovery, and orthographic switching. We find that the highest-scoring open source models output verbatim reference transcript spans even when the relevant audio is contradictory, masked, or ambiguous. Using a variety of mechanistic probes, we show that models respond to narrow acoustic cues to override the faithful representation of the audio in favor of a benchmark-optimized policy. We show the benchmark-optimized behavior can be causally manipulated via low-rank linear steering or simply appending audio to the end of a segment in some cases. Overall, our results indicate that high-performing models exhibit benchmark-conditioned behaviors that can inflate benchmark performance without reflecting improved general-purpose transcription ability.
[AI-24] A Strong Linear Baseline for Whole-Heart Cardiac Shape Completion on CT with an Open Eleven-Structure Statistical Shape Model
链接: https://arxiv.org/abs/2608.19932
作者: Matej Gazda,Jakub Gazda,Juraj Gazda,Peter Drotar
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Public cardiac cohorts annotate different subsets of the heart, so shapes from separate sources cannot be pooled without shared correspondence. Among released cardiac shape resources, none we identified carries the atrial appendage, pulmonary veins, and caval stumps as separate blocks in one mesh. Completion benchmarks also compare deep models against a least-squares projection onto shape modes, not the conditional estimator the same fitted model implies. We release an eleven- structure cardiac computed-tomography (CT) statistical shape model, built from 383 automatically labelled cases in 11 571-vertex correspondence, and compare completion estimators under one frozen internal split and endpoint. On a 76-case internal list held out from fitting, a closed-form conditional-Gaussian estimator reconstructed the missing non-chamber structures at 3.717 mm mean per-vertex error, averaged equally over one, three, five, and nine observed structures. A five-refit mask-conditioned graph variational autoencoder reached 5.248 mm and nearest-neighbour retrieval 8.931 mm. The paired difference was 1.531 mm (95% confidence interval 1.384 to 1.711), and the ordering held in a raw-coordinate sensitivity arm. Expert manual labels exist for 58 external CT cases, but our registered reference is close enough to score only five structures. There the closed-form estimator again had lower average surface distance, 95th-percentile Hausdorff distance, and Chamfer error for both completed atria. On a second public benchmark of 20 cases the reference was close enough for three of four completed structures, and the same ordering held there. Four structures have no expert reference. The released model and its completion operator support cohort-unification research on aligned CT, not clinical use.
[AI-25] Spike-based Belief Propagation in Nonlinear Dynamical Systems
链接: https://arxiv.org/abs/2608.19907
作者: Sepideh Adamiat,Hongye Wang,Wouter M. Kouw,Bert de Vries
类目: Artificial Intelligence (cs.AI); Machine Learning (cs.LG); Neural and Evolutionary Computing (cs.NE); Systems and Control (eess.SY)
备注: Accepted at NCTA 2026 (18th Int’l Conf. on Neural Computation Theory and Applications), part of IJCCI 2026, Angers, France. Pre-peer-review submitted version
Abstract:This paper presents a Bayesian control framework that integrates spike-based dynamics with probabilistic inference for adaptive control. Bayesian inference is widely regarded as a core computational principle of brain function, providing a normative framework for perception, decision-making, and learning under uncertainty. By combining a biologically inspired spiking neural model with Bayesian inference principles, we propose a brain-like control algorithm capable of operating in uncertain environments. We use the mountain car parking problem as a benchmark with non-linear dynamics. Our results demonstrate that the proposed controller can successfully update states in real time and generate goal-directed action plans through spike-driven dynamics. The results highlight the proposed model’s potential as a bridge between computational neuroscience and probabilistic control theory.
[AI-26] Bringing analytic rigor to agent ic AI for science: The Brain Researcher platform for neuroimaging data analysis
链接: https://arxiv.org/abs/2608.19902
作者: Zijiao Chen,Nicholas Lu,Xinhui Li,Jocelyn A. Ricard,Ce Ju,Huan H. Wang,Christian Kindermann,Jeanette A. Mumford,Steven Dillmann,James Kent,Alejandro de la Vega,Sanmi Koyejo,Vince D. Calhoun,Joshua W. Buckholtz,Juan Helen Zhou,Steffen Bollmann,Russell A. Poldrack
类目: Artificial Intelligence (cs.AI); Neurons and Cognition (q-bio.NC)
备注: 103 pages, 19 figures; Supplementary Information included
Abstract:AI agents can execute scientific analyses, but an analytic output becomes a defensible claim only after alternatives are weighed and the claim is limited to what the evidence supports. Agents may reproduce failures including selective analysis, premature declarations of success and optimization of imperfect criteria. We present Brain Researcher, an agentic research harness operating in a neuroimaging researcher’s computational environment under rules for admissible analyses, required checks and claim scope. In benchmarks, Brain Researcher increased first-choice tool-selection accuracy across seven models by 70.2 percentage points (23.3% without it versus 93.6% with it) and verifiable grounding from 4.6% to 22.0%. In collaborator-led and self-evolving studies, multiverse analyses exposed analytic-choice sensitivity, and scientific review classified claims as accepted, qualified, revised, blocked, rejected or deferred. By linking decisions to evidence and provenance, Brain Researcher embeds methodological judgment within the workflow, not after it.
[AI-27] MaliciousSkillBench: A Comprehensive Benchmark for Malicious Agent Skill Detection
链接: https://arxiv.org/abs/2608.19901
作者: Yue Wang,Yi Liu,Gelei Deng,Ying Zhang,Yuekang Li,Zhenyu Chen,Leo Zhang
类目: Cryptography and Security (cs.CR); Artificial Intelligence (cs.AI)
备注: Project page: this https URL
Abstract:Agent Skills extend LLM agents with reusable instruction packages that may also include scripts, resources, and service configuration. This creates a direct distribution channel for malicious behavior, yet existing malicious-Skill datasets are fragmented across sources, artifact formats, evidence regimes, and benign coverage; duplicated and structurally related content further complicates direct aggregation and evaluation. We present MaliciousSkillBench, a comprehensive benchmark for malicious Agent Skill detection. We consolidate 13 public sources, 11 of which contribute Core malicious artifacts, and reduce 8,414 raw malicious records to 7,539 normalized-unique identities in 4,588 operational structural families. After conservative cross-label conflict exclusion, the primary benchmark contains 9,740 Skills: 7,505 malicious and 2,235 benign. To characterize its coverage, we harmonize 11 attack categories for 4,983 malicious identities with supported source-native mappings and find substantial differences in threat composition across sources. We then evaluate three learned text detectors and three off-the-shelf Skill scanners. Learned detectors achieve 0.882-0.932 Random Macro-F1 but only 0.653-0.665 under Source-Disjoint evaluation; the strongest word TF-IDF SVM scores 0.932/0.916/0.665 on Random/structural-disjoint/Source-Disjoint while retaining 95.6% malicious recall but producing 62.4% benign FPR on held-out sources. Off-the-shelf scanners occupy different but also unsatisfactory operating regimes, reducing false positives only at the cost of sharply lower malicious recall. Together, these results show that reliable malicious-Skill detection requires both broader cross-source benchmark coverage and evaluation that jointly measures attack detection and benign over-flagging.
[AI-28] EXIMO: VLM Guided Exploration of VLA Policies
链接: https://arxiv.org/abs/2608.19891
作者: Bhavya Sukhija,Oliver Groth,Mohit Shridhar,Tim Hertweck,Michael Bloesch,Markus Wulfmeier,Abbas Abdolmaleki,Martin Riedmiller
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:How to efficiently finetune robot policies to learn new tasks on the fly? State of the art robotic manipulation policies are based on behaviour cloning of large vision-language-action (VLA) models with billions of parameters on huge teleoperation datasets. While this simple approach has enabled significant advances for robotic manipulation, finetuning of VLA policies for learning new tasks still remains an open problem. In particular, collecting teleoperation datasets requires hundreds of hours of expensive human labour and the alternative, reinforcement learning (RL), can be notoriously sample-inefficient especially for long-horizon tasks. In addition, RL with VLAs imposes several challenges due to the model’s size and architectural design. In this work, we propose EXIMO, an efficient algorithm for finetuning of VLA policies. EXIMO operates in three stages: explore, imitate, and optimize. During the explore phase, EXIMO equips the VLA with a vision language model (VLM) that acts as a planner. The VLM thinks and breaks down challenging long-horizon problems into shorter ones for the VLA. The VLM, together with the VLA, is used to collect an orchestrated dataset on new tasks. During the imitate phase, the VLA is finetuned with the orchestrated data. Finally, during the optimize stage, we use residual off-policy RL to further finetune the policy. In our experiments, we ablate all three stages of EXIMO and show that it outperforms existing approaches significantly in terms of sample-efficiency and final performance.
[AI-29] Write Once Run Everywhere: The Axon DSL for Shape-Safe and Framework-Agnostic LLM Architectures
链接: https://arxiv.org/abs/2608.19889
作者: Jacob Nielsen,Danial Namazifard,Lukas Galke Poech,Peter Schneider-Kamp
类目: Artificial Intelligence (cs.AI); Programming Languages (cs.PL)
备注:
Abstract:The entire ecosystem of open-source language models effectively relies on a single platform. What if this platform was forced to shut down tomorrow? Implementing and maintaining efficient model definitions and translating them between different training and inference regimes is a resource-heavy task that severely limits model efficiency and portability, hindering both scaling and deployment. Here, we present Axon, a strongly typed domain-specific language with Haskell-like syntax, that enables a write-once, run everywhere paradigm for LLM architectures. By basing collaboration on a language specification rather than a specific framework’s vision, Axon fosters open cooperation and empowers researchers to implement highly specialized architectures without giving up optimization infrastructure or accepting deployment lock-in. Axon allows for concise, auditable specifications that can be automatically compiled to standalone implementations for leading frameworks: PyTorch, PyTorch with Triton, JAX, MLX and vLLM. In 467 inference benchmarking experiments on models ranging from 135M to 32B parameters, we demonstrate median speedups of 7% on PyTorch, 12% on PyTorch with Triton, 91% on JAX, and 107% on MLX, compared to the reference implementations from Transformers. When deployed as native vLLM architectures with PagedAttention and KV-cache, Axon models achieve a 58% median speedup over Transformers implementations.
[AI-30] Evidence Before Expansion: Reuse Spawn or Defer in Lifelong Expert Pools
链接: https://arxiv.org/abs/2608.19888
作者: Kentaro Oda
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Neural and Evolutionary Computing (cs.NE)
备注: 8 pages, 4 figures
Abstract:Streaming systems that maintain a pool of expert models must repeatedly decide whether to reuse an existing expert for arriving data, spawn a new one, or defer. We present a decision layer that makes all three outcomes statistically meaningful. Reuse and spawn are posed as one-sided sequential hypotheses on a conditional (mechanism-level) discrepancy, separated by an indifference zone; defer is exactly the state in which neither betting e-process has accumulated sufficient evidence. We prove finite-time anytime validity for the observable surrogate discrepancy of a predictable discriminator sequence, and an unconditional one-sided transfer to the population quantity in which each side’s slack is the excess risk of a single discriminator; an empirically observed downward-bias regularity makes the spawn side exactly conservative. Recency without sacrificing the guarantee is obtained by a restarted e-detector: a bank of unwindowed betting supermartingales at geometrically spaced restart times (O(log t) memory), with the error budget spent over restart instances, which preserves lifetime anytime validity; spending over expert-creation order likewise controls multiplicity for unboundedly many experts. On synthetic multi-concept streams, Electricity, Covertype, and the recurrence-heavy INSECTS benchmark, the instance-accounted restarted bank achieves zero false spawns and zero false reuses after switches and matches or exceeds the retired windowed heuristic (INSECTS-reoccurring accuracy 0.675), making the deployed algorithm and the guaranteed algorithm one and the same.
[AI-31] Separating Covariate Shift from Mechanism Change with Two Discriminators: CJSD a Conditional Discrepancy with an Exact Covariate-Concept Decomposition
链接: https://arxiv.org/abs/2608.19885
作者: Kentaro Oda
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
备注: 11 pages, 7 figures
Abstract:Streaming systems that maintain a pool of expert models must repeatedly decide whether to reuse an existing expert for arriving data, spawn a new one, or defer. We present a decision layer that makes all three outcomes statistically meaningful. Reuse and spawn are posed as one-sided sequential hypotheses on a conditional (mechanism-level) discrepancy, separated by an indifference zone; defer is exactly the state in which neither betting e-process has accumulated sufficient evidence. We prove finite-time anytime validity for the observable surrogate discrepancy of a predictable discriminator sequence, and an unconditional one-sided transfer to the population quantity in which each side’s slack is the excess risk of a single discriminator; an empirically observed downward-bias regularity makes the spawn side exactly conservative. Recency without sacrificing the guarantee is obtained by a restarted e-detector: a bank of unwindowed betting supermartingales at geometrically spaced restart times (O(log t) memory), with the error budget spent over restart instances, which preserves lifetime anytime validity; spending over expert-creation order likewise controls multiplicity for unboundedly many experts. On synthetic multi-concept streams, Electricity, Covertype, and the recurrence-heavy INSECTS benchmark, the instance-accounted restarted bank achieves zero false spawns and zero false reuses after switches and matches or exceeds the retired windowed heuristic (INSECTS-reoccurring accuracy 0.675), making the deployed algorithm and the guaranteed algorithm one and the same.
[AI-32] ESTNAV: Pareto-Guided Search for Compositional Robustness Testing
链接: https://arxiv.org/abs/2608.19882
作者: Arooj Arif,Tobias Hartung,Elena Botoeva,Alexandros Koliousis
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Deep learning models remain vulnerable to real-world input perturbations, especially when multiple corruptions co-occur in the same input (e.g., brightness shifts and motion blur). Compositional testing reveals these interaction effects but introduces two challenges: combinatorial growth of the perturbation space as dimensions and severity levels increase, and uneven diagnostic value-many combinations yield unrealistically degraded inputs with limited practical relevance. We present TESTNAV, 1 a Pareto-guided robustness testing framework for efficiently exploring discrete, compositional perturbation spaces when only a limited number of perturbation configurations can be evaluated. TESTNAV prioritises severe yet realistic failures by formulating robustness testing as bi-objective optimisation: maximise performance degradation while preserving input fidelity measured by modality-specific metrics (e.g., SSIM and KID for vision; chrF and BERT-F1 for language and code). It uses NSGA-II to approximate the bi-objective Pareto front. Across four benchmarks spanning vision, natural language, and code generation, TESTNAV recovers Pareto fronts up to 2.15x faster than search-based baselines, using 35.8%-89.3% of the discrete perturbation space defined by four perturbation dimensions with six levels each. Subjects: Artificial Intelligence (cs.AI) Cite as: arXiv:2608.19882 [cs.AI] (or arXiv:2608.19882v1 [cs.AI] for this version) https://doi.org/10.48550/arXiv.2608.19882 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
[AI-33] Repo0: Design-Driven Zero-to-All Code Generation
链接: https://arxiv.org/abs/2608.19854
作者: Silin Chen,Haoyi Teng,Xiaodong Gu,Yuling Shi,Jiale Huang,Yongpan Wang,Hongyu Zhang,Haibing Guan
类目: oftware Engineering (cs.SE); Artificial Intelligence (cs.AI)
备注: Our code and data are available at this https URL
Abstract:Large language model agents have made substantial progress in code generation, yet most existing systems assume a predefined repository architecture. This assumption does not hold in zero-to-all code generation, where an agent must construct an entire software project directly from natural-language requirements while maintaining a modular repository architecture throughout development. We present Repo0, a continuous structural evolution framework for zero-to-all code generation. Repo0 maintains an explicit architectural state instantiated as a Dual-Directed-Acyclic-Graph (Dual-DAG), consisting of a requirement-level DAG, a component-level DAG, and their alignment relation. Starting from natural-language requirements, it iteratively evolves component boundaries through structural actions guided by modularity metrics until structural convergence, after which the converged architecture guides test-driven development code generation. We evaluate Repo0 on six real-world repositories from RepoCraft using GPT-5 mini and DeepSeek V3.2. Repo0 achieves the highest Functionality Coverage and Pass Rate across all settings. Compared with RPG, the strongest repository-planning baseline, Repo0 improves Functionality Coverage by up to 20.08 percentage points and Pass Rate by up to 29.74 percentage points. Ablation and structural-evolution analyses further demonstrate the importance of the Dual-DAG architectural state, modularity-guided structural evolution, and explicit structural convergence.
[AI-34] SAPO: Single-Rollout Autoregressive Policy Optimization for Agent ic Reinforcement Learning
链接: https://arxiv.org/abs/2608.19842
作者: Dayang Liang,Lang Feng,Bo An,Yunlong Liu
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Agentic reinforcement learning (RL) has become a critical stage in the post-training of large language models. Existing critic-free, group-relative methods estimate policy advantages from multiple rollouts, avoiding the substantial memory overhead of conventional proximal policy optimization (PPO) and achieving strong performance on long-horizon interactive tasks. Despite their success, recent studies revealed three limitations: (1) Lack explicit value generalization and effective temporal credit assignment; (2) Suffer from potential advantage collapse in long-horizon complex tasks; (3) Require a costly trade-off between sampling budget and policy performance. In this work, we propose Single-rollout Autoregressive Policy Optimization (SAPO), a low-memory and compute-efficient framework in which the policy and value functions share a single autoregressive backbone. SAPO exploits the autoregressive structure of LLMs to produce policy and value predictions at distinct causal boundaries with shared parameters, while independently optimizing the PPO objectives and auxiliary on-policy SARSA objectives. To robustly estimate the contribution of each turn, we further introduce a trajectory-level generalized advantage estimator that combines lambda-returns with batch normalization. Experiments across ALFWorld and WebShop with Qwen2.5-1.5B/7B show that SAPO trains stably and outperforms PPO and GRPO by mean +15.1 and +12.1 percentage points, respectively, while eliminating the memory cost of a separate critic model and reducing per-iteration runtime by 33.2% over PPO.
[AI-35] Specification-delta-driven data governance: an empirical study of the «spec-delta» as the unit of change in lakehouse data platforms
链接: https://arxiv.org/abs/2608.19838
作者: Pablo Ramirez Amador
类目: Artificial Intelligence (cs.AI); Programming Languages (cs.PL)
备注:
Abstract:Spec Driven Development SDD has consolidated the idea that the specification rather than the code should be the primary artefact governing AI assisted work. Tools such as GitHub Spec Kit, and proposals such as Constitutional SDD, have formalised this principle in the software domain, while the executable data-contracts literature has extended it to schema and quality enforcement at run time. Nevertheless, the treatment of the specification delta OpenSpec’s core idea that every change should produce a reviewable increment of requirements as the unit of change in data platforms remains empirically unexplored, even though many data-platform changes are contractual (new datasets, service-level agreements, metric semantics, access policies) rather than purely code changes. This work formalises the spec-delta concept, proposes a taxonomy of data platform changes according to their suitability for incremental specification, and defines a controlled experiment comparing a spec-delta-driven workflow against a conventional code pull-request workflow without a delta. The response variables are discovery to deployment time, the density of defects reaching the Silver and Gold lakehouse layers, cross-tool metric divergence, and reviewer cognitive load measured with NASA TLX. The paper explicitly reserves a demonstration-and-laboratory section for instantiation on a real lakehouse environment. The contribution is not a tool but reproducible evidence and an applicability guide that helps to avoid the up front over specification antipattern.
[AI-36] Adaptive Probabilistic Shielding by Learning MDPs for Safe Reinforcement Learning
链接: https://arxiv.org/abs/2608.19836
作者: Astrid Horn Brorholt(1),Maris F. L. Galesloot(2),Nils Jansen(2) (3),Kim Guldstrand Larsen(1),Christian Schilling(1) ((1) Aalborg University, Aalborg, Denmark (2) Radboud University, Nijmegen, Netherlands (3) Ruhr University Bochum, Bochum, Germany)
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Logic in Computer Science (cs.LO)
备注: 19 pages, 3 figures, 3 tables. To be published in the proceedings of RV 2026
Abstract:Probabilistic shielding is a technique for safe reinforcement learning (RL). Typically, a static observer – called the shield – constrains the learning agent’s actions to those for which acting safely remains feasible. Traditionally, the shield is computed from the transition probabilities of the underlying Markov decision process (MDP). Thus, this technique is not applicable when the MDP model is not given a priori, which, unfortunately, is the case in typical RL applications. In this paper, we study the problem of computing a shield in the setting where the transition graph of the MDP is known, but the transition probabilities are unknown. Our approach integrates probabilistic shielding with online model learning: as the RL agent explores the environment, we estimate the transition probabilities. From this estimate, we compute a shield. While the shield may be conservative initially, it adapts as the model estimate becomes more precise. Thus, the shield improves in tandem with the RL agent. This paradigm of adaptive probabilistic shielding raises a number of challenges, such as when to recompute the shield and how to balance between exploration and safety during learning. We empirically evaluate multiple variants of this paradigm across several environments.
[AI-37] Causal Reasoning with Bipartite Graphical Causal Models UAI
链接: https://arxiv.org/abs/2608.19831
作者: Joris M. Mooij
类目: Artificial Intelligence (cs.AI); Probability (math.PR)
备注: Figure 2 was added w.r.t. to the published version at Uncertainty in Artificial Intelligence (UAI) 2026
Abstract:Causal Bayesian networks (CBNs) and structural causal models (SCMs) are the dominant frameworks for graphical causal reasoning, but they cannot adequately represent all real-world causal systems. In particular, systems at equilibrium—where feedback mechanisms create cyclic causal dependencies—can exhibit causal semantics that are fundamentally incompatible with these frameworks: different interventions that enforce the same variable value may have different effects, rendering the standard ``perfect intervention’’ do( X = x ) ambiguous. We propose bipartite graphical causal models (BGCMs), in which the structure of a system of equations is encoded by a bipartite graph with variable and equation nodes. In this framework, a hard intervention do( f_j : X_v = \xi_v ) specifies which equation is replaced, which variable is targeted, and at what value—resolving the ambiguity of the standard notion. We demonstrate, through a detailed case study of a physical system, that this representation naturally corresponds to distinct real-world interventions. We formulate a Markov property in terms of a new graphical separation criterion (B-separation) that exploits the functional determinism inherent in the equations, and we extend it to settings with non-random inputs. We show how this gives rise to a do-calculus for reasoning about domain invariances. BGCMs strictly generalize CBNs and SCMs while retaining the ability to perform graphical causal reasoning.
[AI-38] ADAPT: Physics-Aware Diffusion-based World Models for Adaptive Predictive Transferable HVAC Control
链接: https://arxiv.org/abs/2608.19804
作者: Xu Yang,Kailai Sun,Dianyu Zhong,Qianchuan Zhao
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Buildings account for roughly one-third of global energy consumption and CO _2 emissions. Optimizing indoor climate systems plays a critical role for urban climate mitigation aligned with UN Sustainable Development Goals 11 and 13. However, indoor delayed thermodynamic responses and partial observability severely hinder existing methods, which are primarily limited by implicit thermal inertia, occupancy dynamic prediction, and cumulative prediction errors, especially for out-of-distribution environments. In practice, these challenges are further exacerbated by the high cost and privacy burden of dense indoor sensing, forcing operators to collect only limited data in a single operating regime while expecting controllers to generalize reliably across unseen seasons and climate regions. To address this problem, we propose ADAPT, a physics-aware conditional diffusion indoor environmental world model for HVAC control. The model predicts a short-horizon held-action thermal baseline to capture the latent thermal inertia of the buildings. The diffusion backbone utilizes the robustness of generative models, while a learnable multi-zone heat-balance regularizer constrains generated trajectories to satisfy transferable building thermodynamics without requiring known building geometry or manually calibrated thermal parameters. A credit assignment is then design for the downstream reinforcement learning. Extensive experiments on SemibuildingSim and Sinergym demonstrate that ADAPT reduces HVAC energy consumption by 7.3% and occupant discomfort by 30.2% compared with state-of-the-art baselines under IID control. Under OOD control scenarios spanning unseen seasons and climate regions, ADAPT maintains robust performance with only marginal degradation relative to its IID performance, substantially outperforming existing methods in transfer robustness.
[AI-39] owards general embodied intelligence: integrating large language models knowledge bases and reasoning capabilities to build the next generation of AI agents
链接: https://arxiv.org/abs/2608.19794
作者: Fujiang Yuan,Xia Huang,Lusheng Wang,Jun Ding,Zhen Tian,Yuxin Wang,Shaojie Gu,Yuki Funabora,Yanhong Peng,Zebing Mao
类目: Artificial Intelligence (cs.AI); Robotics (cs.RO)
备注:
Abstract:The convergence of large language models (LLMs), structured knowledge bases (KBs), and reasoning ability (RA) presents a promising trajectory toward general embodied intelligence (GEI). This paper reviews the evolution of LLM-centered intelligent systems, emphasising their integration with knowledge representation, logical reasoning, and physical embodiment. We analyse LLM architectures, pre-training methods, and inference mechanisms, along with their interaction with external knowledge sources and structured reasoning frameworks. Furthermore, we examine embodied intelligence (EI) paradigms wherein agents learn and act in physical environments. To synthesise these dimensions, we present a conceptual framework that illustrates the synergy among LLMs, KBs, RA, and embodiment, serving as a guiding model for perception, reasoning, and action rather than an implemented engineering architecture. To advance toward GEI, we identify five key challenges: efficient LLM deployment, closed-loop knowledge integration, hybrid symbolic-neural reasoning, perception-action grounding, and continual learning. This survey provides a comprehensive roadmap for developing adaptive, multimodal agents capable of operating in complex, dynamic settings.
[AI-40] LLM s as Acquisition Policies for Finite-Pool Materials Optimization: A Controlled Study
链接: https://arxiv.org/abs/2608.19790
作者: Dino-Rober Demir,Florian Le Bronnec,Rio Yokota
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Discovering materials with desirable properties often requires searching large candidate spaces while experimental or computational evaluations remain costly. Active learning addresses this challenge by using previous observations to select which candidate to evaluate next, typically through probabilistic surrogate models. We investigate whether open-weight large language models (LLMs) can serve as standalone acquisition policies in this setting. We evaluate five LLMs across four retrospective finite-pool materials optimization tasks under different candidate-presentation strategies and compare them with random selection and conventional Gaussian-process methods. LLM policies generally reach the global optimum in fewer iterations than random selection, indicating that they provide a useful acquisition signal without task-specific training. Their performance relative to Gaussian-process methods is mixed: conventional acquisition performs better on most tasks, while LLMs match or outperform it in some settings. Performance varies substantially across tasks, models, initializations, and candidate presentations, with no LLM approach performing best across all tasks. Overall, open-weight LLMs show potential as acquisition policies for finite-pool materials search, although their reliability remains sensitive to the task and to how candidates and scientific context are presented.
[AI-41] -net: Quantum Inspired Tensor Network Denoising in Conditional GANs
链接: https://arxiv.org/abs/2608.19789
作者: Michal A. Sterzel,Marko J. Rančić
类目: Artificial Intelligence (cs.AI); Quantum Physics (quant-ph)
备注:
Abstract:Developed as a workhorse for classical simulations of quantum algorithms and quantum many-body systems, Tensor Network methods have entered the scientific mainstream in quantum physics. Among various types of tensor networks, Tensor Trains (commonly know as Matrix Product States in the quantum computing community) have already found applications in machine learning. These methods often rely on a powerful linear algebra tool called the Singular Value Decomposition (SVD). Several conditional GAN architectures for image denoising incorporate SVD as a single-cut decomposition step applied to generator feature maps. In this work we introduce TT-Net, which replaces the per-channel SVD denoising block with a two-cut tensor-train decomposition capable of accessing cross-channel information directly, a capability absent from contemporary alternatives. In a controlled comparison differing only in this decomposition mechanism, TT-Net outperforms SVD-Net on PSNR and SSIM across all three noise types tested (Gaussian, motion blur, and salt-and-pepper), supporting the hypothesis that cross-channel access improves denoising quality. Training-dynamics analysis further shows that TT-Net’s adversarial loss term consistently saturates to a stagnant state across all three noise types, more so than SVD-Net’s, while reconstruction quality continues to improve regardless, raising an open question about the adversarial component’s contribution that this work identifies but does not resolve. Furthermore, for Gaussian noise our method outperforms both the EigenGAN and the state of the art Pix2pix method which does not assume any linear algebra decompositions and does not retain any linear algebra information. Our manuscript shows how quantum inspired tools can be used as practical real world feature filters for deep learning applications.
[AI-42] CoToGrasp: Contact-Topology-Conditioned Dexterous Grasp Synthesis via Canonical Workspace Learning
链接: https://arxiv.org/abs/2608.19776
作者: Julien Merand,Boris Meden,Liming Chen,Mathieu Grossard
类目: Robotics (cs.RO); Artificial Intelligence (cs.AI)
备注: Project website: this https URL
Abstract:Current dexterous grasp planners primarily optimize for physical stability, focusing on whether an object can be grasped rather than how it should be grasped to support downstream functional tasks. However, conditioning grasp synthesis on specific human grasp taxonomies typically requires prohibitively expensive, object-annotated datasets. To address these limitations, we propose CoToGrasp, a novel generative framework that synthesizes diverse, stable grasps strictly conditioned on specific contact topologies. To bypass the data collection bottleneck, CoToGrasp is trained entirely in an object-agnostic manner. We introduce a feature-based canonical workspace that projects local object features into a unified gripper-centric domain, effectively decoupling the semantic functional intent from the arbitrary object geometry. By learning the intrinsic contact manifold of the gripper within this workspace, our model achieves zero-shot generalization to unseen objects at inference. Extensive evaluations on the large-scale DexGraspNet dataset demonstrate that CoToGrasp achieves state-of-the-art performance, outperforming existing taxonomy-guided planners. Finally, we demonstrate the physical viability and kinematic feasibility of our synthesized contact topologies on a physical robot platform. Code is available on our project website this https URL .
[AI-43] Finite-Horizon Input-Output Dynamics of Minibatch Perturbations in AdamW
链接: https://arxiv.org/abs/2608.19762
作者: Kang Liu,Suyan Li
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Optimization and Control (math.OC); Machine Learning (stat.ML)
备注:
Abstract:A minibatch can influence training beyond the update at which it is observed because AdamW stores past gradient information in its optimizer states. We study this delayed effect through paired trajectories that differ only in one gradient update and share the same subsequent training sequence. We formulate AdamW as a finite-horizon input–state–output (ISO) system whose state contains the model parameters and first- and second-moment estimates. Linearizing the joint dynamics yields a signed response operator that maps a localized gradient perturbation to its future loss effects, revealing how optimizer memory shapes their magnitude, timing, and sign. We further derive an exact multistep error decomposition and establish first-order finite-horizon accuracy under local smoothness and controlled activation switching. Experiments validate the response mechanism and optimizer-state effects, while repeated-future analyses reveal substantial prospective structure in delayed influence that can be partially recovered from ISO approximations. Code is available at this https URL.
[AI-44] GOAG: Generative and Object-Agnostic Grasp Planner for Dexterous Robotic Manipulation
链接: https://arxiv.org/abs/2608.19759
作者: Julien Merand,Boris Meden,Mathieu Grossard,Liming Chen
类目: Robotics (cs.RO); Artificial Intelligence (cs.AI)
备注: Project website: this https URL
Abstract:Multifingered grasping is a crucial robotic skill, but current deep-learning grasp planners often struggle to generalize to new objects because they are trained on limited, object-specific datasets. We introduce a fundamentally different approach, grounded in the observation that the gripper and the object share identical surface geometry at their mutual contact points. We propose GOAG: Generative and Object-Agnostic Grasp Planner for Dexterous Robotic Manipulation, a novel deep generative model that learns a compact latent representation of a specific gripper’s contact surface distribution, enabling the efficient sampling of valid grasp configurations without relying on object-specific training data. We show that by introducing object features only at inference time, our model can effectively retrieve admissible contact areas that are compatible with the gripper’s capabilities. We validate our approach through extensive experiments on established grasp protocols in both simulated and real-world scenarios, demonstrating its effectiveness with different grippers from the literature. Our method delivers state-of-the-art results on the objects from the MultiDex dataset, achieving an average success rate of 86.93%. It offers significantly faster processing when generating numerous grasps, while matching the performance of leading approaches specifically trained on this dataset. Unlike these methods, our approach does not rely on object-specific training data, highlighting the advantages of object-agnostic learning. It effectively addresses the generalization challenges faced by traditional data-driven grasp planners. Code and videos are available on our project website this https URL .
[AI-45] GenMatch: An End-to-End Generative Matching Framework for Micro-View Order-Dispatching in Ride-Hailing
链接: https://arxiv.org/abs/2608.19751
作者: Chuang Liu,Yuxueqing Zhang,Tengfei Lyu,Zirui Yuan,Weiqi Hu,Yanghan Cheng,Ming Wang,Li Ma,Zihao Lu
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Micro-View Order-Dispatching assigns available drivers to passenger orders within each dispatch batch and is critical to the service quality and operational efficiency of ride-hailing platforms. Mainstream industrial solutions follow a multi-stage paradigm of model prediction, value calculation, and dispatch matching. Although dispatch quality is determined by the final batch-level assignment, these stages optimize different intermediate objectives. This cross-stage objective inconsistency means that improving a single stage does not necessarily improve the overall dispatch result. We therefore formulate Micro-View Order-Dispatching as a generative matching problem and propose GenMatch, an end-to-end Generative Matching framework and the first such framework deployed in a real-world production environment. Applying generative modeling to this problem introduces three challenges. First, each dispatch batch forms a dynamic sparse bipartite graph, requiring efficient structured batch-level encoding. Second, replacing the hand-crafted value function requires learning unified business utility from heterogeneous feedback. Third, directly generating an assignment requires tracking the evolving matching state because each selected order-driver pair changes the remaining feasible candidates. GenMatch addresses these challenges with a Context-Aware Bipartite Encoder, a Business-Aware Utility Learner, and a State-Aware Pointer Decoder. Extensive offline evaluations and online A/B tests in five cities across DiDi’s international ride-hailing markets show consistent improvements over competitive baselines, confirming the effectiveness and practicality of GenMatch for industrial order-dispatching.
[AI-46] A Locally Tokenized Generative Model for Robust Time-Series Watermarking NEURIPS2026
链接: https://arxiv.org/abs/2608.19727
作者: Dongbin Kim,Geonwoo Shin,Yujin Choi,Soyeon Park,Jaewook Lee
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
备注: Submitted to NeurIPS 2026
Abstract:Watermarking is a central tool for provenance in generative models, yet its application to multivariate time series remains hindered by reliability failures under post-editing attacks. We show that existing detectors, which rely on globally coupled re-encoding, suffer from bidirectional drift of the null distribution: post-editing attacks can shift the z-score of non-watermarked samples in either direction, invalidating clean-calibrated thresholds. We argue that this instability is a property of the re-encoding, and that reliable detection requires each recovered unit to depend only on a bounded temporal neighborhood. Guided by this principle, we propose L-VQVAE, a generative model in which each discrete token is produced from a short contiguous window, and LVQMark, a watermarking method over this token space that combines logit-bias injection with robust re-encoding for attack-time detection. Experiments on four benchmarks spanning finance, energy, and neuroimaging show that our approach preserves generation quality while stabilizing both detection power and false-positive behavior under post-editing attacks.
[AI-47] Loreley: Repository-Scale Program Evolution with Quality-Diversity Search CEC
链接: https://arxiv.org/abs/2608.19703
作者: Mohan Chen
类目: oftware Engineering (cs.SE); Artificial Intelligence (cs.AI)
备注: 15 pages, 3 figures, 8 tables. Code and evidence: this https URL
Abstract:Sequential agent search accumulates changes from its current champion but discards alternative branches; independent proposals preserve breadth but restart from the root. Loreley instead retains complete repository states in a Quality-Diversity (QD) archive and samples them as parents or supplies them as context for later edits. Candidates are Git commits produced in isolated worktrees and judged by a project-supplied evaluator. We compare configured Loreley QD, sequential champion editing, and independent root proposals in a matched Zstandard experiment: seven paired blocks and 48 physical candidate jobs per policy and block (1,008 total), with root-only initialization and each policy’s native concurrency. Validation selected a winner at each budget checkpoint; an agent-hidden holdout measured the fixed candidate. At 48 jobs, QD was 0.135% below Sequential Champion (95% BCa interval for the paired effect: -0.556% to +0.161%) and 0.320% above Independent Root (-0.082% to +0.686%). Neither contrast established a QD advantage; Sequential had the highest observed 48-job mean and median. Archive retention and later sampling did occur. Four of seven final QD winners had a non-incumbent state in their primary-parent ancestry under a retrospective one-incumbent rule applied only to the observed QD stream. Including inspiration edges raised the count to six, without showing that supplied context caused an edit. Three earlier capability campaigns produced generation-4, multi-file improvements in two Python libraries and a separate Zstandard revision. Loreley engaged the intended stepping-stone mechanism, but the controlled experiment did not show an endpoint benefit at 48 jobs.
[AI-48] Beyond Memory Majority: Latent-Source Reasoning for Multi-Agent Memory Arbitration
链接: https://arxiv.org/abs/2608.19701
作者: Chenchen Lin,Wenhao Yuan,Xuehe Wang,Edith Cheuk Han Ngai
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Long-term multi-agent systems continuously accumulate the memories produced by different agents. Existing memory methods typically treat retrieved memories as independent evidence and combine them through voting or weighting. However, this independence assumption often fails in multi-agent settings: memories written by different agents may inherit the same upstream source or shared bias, causing correlated evidence to be repeatedly counted and creating a false majority. We term this failure mode \textitMemory Correlation Bias. To address the issue, we propose the \textbfCorrelation-\textbfAware \textbfMemory \textbfArbitration (CAMA) framework that jointly decouples retrieved memories and recovers missing independent evidence. We model the retrieved memories as query-conditioned evidence groups and combine neural dependency inference with provenance-based symbolic priors to estimate the effective number of independent evidence sources, thereby preventing correlated memories from forming a false majority. Since critical independent evidence may be absent from the initial retrieval set, \textscCAMA further learns a sequential recovery policy that actively retrieves alternative evidence or traces upstream sources before making the final decision, aiming to recover sufficient independent evidence for reliable arbitration while minimizing retrieval cost. Experiments on multiple benchmarks demonstrate the superiority of our method over the state-of-the-art baseline methods, suppressing false majorities induced by correlated memories.
[AI-49] Rethinking the Evaluation and Optimization of LLM -Based Social Simulation
链接: https://arxiv.org/abs/2608.19689
作者: Pei Wang,Xu Chen,Ji-Rong Wen
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:LLM-based social simulation is a promising complement to traditional methods such as surveys and behavioral experiments. A core question is how to evaluate the fidelity of LLM-simulated human behavior and optimize LLMs toward it. Prevailing practice evaluates by accuracy, checking whether the model selects the single response observed from a human, and trains the LLM to reproduce this hard label. However, human behavior is inherently subjective: the same person in the same situation may reasonably act differently, so an observed response is only one draw from an underlying response distribution, rendering accuracy-based evaluation unreliable and hard-label training misleading. To address these problems, we first introduce the subjectivity coefficient, an entropy-based quantity distinguishing objective tasks such as coding from subjective ones such as social simulation, and use it to systematically analyze how accuracy-based evaluation and hard-label training fail as subjectivity grows. Based on the subjectivity coefficient, we propose Subjectivity-Adaptive soft-Label Training (SALT): it pools observed outputs from semantically nearby inputs into soft distributional labels, with an aggregation radius adapted to the estimated subjectivity of each input; in the near-objective limit the neighborhood shrinks, so SALT naturally falls back to standard single-label training. Moreover, since existing datasets record only single observed responses and cannot support distributional evaluation, we construct SUBJSIM, a benchmark of 19,300 contexts covering 193 annotators and 100 subjective questions. Since real-world data typically provide only a single observation per input, our experiments train models from single observed outputs while evaluating them against the full response distributions, verifying feasibility in realistic settings. Results on SUBJSIM demonstrate the advantages of our method.
[AI-50] Learning Hierarchical Skill Policies with Offline Quality-Diversity Reinforcement Learning IROS2026
链接: https://arxiv.org/abs/2608.19684
作者: Tanachai Anakewat,Takayuki Osa,Tatsuya Harada
类目: Artificial Intelligence (cs.AI); Machine Learning (cs.LG); Robotics (cs.RO)
备注: IROS 2026
Abstract:Recent studies investigate how to leverage pre-collected datasets to improve the policy performance and sample efficiency of RL. One promising approach to achieve this goal is to employ a two-stage strategy: In the first stage, diverse skills are extracted as a low-level policy from a given dataset, and a high-level policy is trained to solve a specific task in the second stage. Typically, extraction of the low-level policy is performed based on unsupervised learning such as trajectory VAE. However, a limitation of this approach is that the quality of the low-level policy highly depends on the quality of the dataset. To address this issue, we introduce QDOS (Quality-Diversity Offline Skill learning), a unified pipeline for robust offline-to-online learning. Our approach incorporates an Advantage-Weighted Quality-Diversity pretraining objective, which weights the skill extraction and diversity objectives by the estimated advantage of each trajectory segment. This approach allows the model to extract diverse and high-value skills. By providing robust and task-relevant skill representations, QDOS significantly improves the quality of the embedded skill space used by the low-level policy. We further integrate this with a dual dataset reuse strategy, where offline data is used both for skill pretraining and for populating the online replay buffer via pseudo-labeling. Experiments demonstrate that QDOS significantly outperforms strong baselines in structured manipulation tasks and unstructured locomotion tasks, confirming its ability to accelerate exploration and improve final returns in challenging sparse-reward domains.
[AI-51] Frequency-Aware Continual Learning for Smart Contract Vulnerability Detection with Large Language Models
链接: https://arxiv.org/abs/2608.19680
作者: Tenghui Huang,Jiawen Kang,Dongning Liu,Changyan Yi,Chengjun Cai,Anjia Yang,Li Li,Dong In Kim
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Smart contract vulnerability detection with Large Language Models (LLMs) faces three causally linked challenges. First, new vulnerability categories demand parameter-efficient adaptation, since full retraining is prohibitive for sequentially arriving tasks. Second, training per-task adapters on a shared backbone causes catastrophic forgetting of previously learned vulnerabilities. Third, the resulting multiplicity of adapters must be consolidated into a single model, since task identity is unknown at inference time. Each challenge arises directly from the solution to its predecessor, making an integrated framework essential. We propose a three-stage pipeline in which each stage addresses one challenge and feeds into the next. The adaptation stage uses Frequency-Aware Low-Rank Adaptation (FA-LoRA), which performs adaptation in the Fourier domain with per-frequency importance gates, requiring only 0.4% trainable parameters while outperforming standard LoRA and QLoRA. The continual learning stage applies Forget-Aware Replay (FAR), which uses these frequency gates to estimate per-sample forgetting risk via loss dynamics and prioritizes vulnerable knowledge for rehearsal, achieving an average Micro-F1 of 0.8022 across sequential tasks. The deployment stage employs Anchor-Protected Progressive Merging (APPM), which exploits the asymmetric generalization produced by FAR training to identify the strongest-generalizing adapter as an anchor and consolidates all adapters into a single model via anchor-protected weighted merging with frequency-domain gate competition. APPM achieves a Micro-F1 of 0.8085, within 2.7% of the independent per-task upper bound, at a merge cost of 156 ms and no additional runtime memory. Experiments on DIVE confirm the framework effectively addresses all three challenges for evolving blockchain ecosystems.
[AI-52] Escaping the Quicksand: A Call to Arms
链接: https://arxiv.org/abs/2608.19674
作者: Peter Sewell,Jean Pichon-Pharabod
类目: Programming Languages (cs.PL); Artificial Intelligence (cs.AI); Software Engineering (cs.SE)
备注:
Abstract:Computing has been an astonishing success - but the accumulated technical debt exposes us all to huge costs in business and societal risk. For 75 years, we’ve built systems to prose specifications with test-and-debug development. That works well enough for industry to thrive, but it’s an expensive and ineffective feedback loop, and leaves everyone relying on shaky foundations. Now, AI-enabled engineering is amplifying the success by reducing coding costs, but also amplifies the risks, by rapidly increasing technical debt, and by automating detection of the vulnerabilities therein. How can we do better? Research has long pursued mathematical proof of correctness, which, unlike testing, can cover all cases. This too has advanced massively, but it remains hard to apply, both technically and because of a deep-seated cultural disconnect. Instead, we argue for a pragmatic approach to flexible combinations of testing, specification, and proof, that provides more effective feedback loops for both AI and human development. Most simply, one can incrementally co-develop executable-as-test-oracle partial specifications alongside conventional prose descriptions, code, and tests. This clarifies design and makes testing much more discriminating. Developers can and should do it today. Or, even better, one can use specifications that support the full gamut of testing, property-based testing, symbolic execution, and proof. This enables a range of intertwined feedback loops, again both for AI and humans, from cheap testing to more expensive proof. However, making it really practical needs semantics infrastructure: specifications and tooling for the main programming languages and other abstractions, which we now more-or-less know how to build, but which is not yet in place. We call the community to arms to create and deploy it - to enable a future built on firmer ground. Subjects: Programming Languages (cs.PL); Artificial Intelligence (cs.AI); Software Engineering (cs.SE) Cite as: arXiv:2608.19674 [cs.PL] (or arXiv:2608.19674v1 [cs.PL] for this version) https://doi.org/10.48550/arXiv.2608.19674 Focus to learn more arXiv-issued DOI via DataCite (pending registration) Submission history From: Peter Sewell [view email] [v1] Thu, 20 Aug 2026 06:07:56 UTC (71 KB) Full-text links: Access Paper: View a PDF of the paper titled Escaping the Quicksand: A Call to Arms, by Peter Sewell and Jean Pichon-PharabodView PDFHTML (experimental)TeX Source view license Current browse context: cs.PL prev | next new | recent | 2026-08 Change to browse by: cs cs.AI cs.SE 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
[AI-53] DeltaML-Bench: Evaluating Machine Learning Agents on Real-World Research Repositories
链接: https://arxiv.org/abs/2608.19653
作者: Josias Moukpe,Priyanka Aryal,Matthew Kenney
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
备注: 18 pages, 3 figures, 12 tables. Code and benchmark: this https URL
Abstract:Autonomous agents for machine learning experimentation must navigate heterogeneous repositories, repair training pipelines, and evaluate candidate improvements under realistic compute constraints. Existing benchmarks only partially capture these conditions. We introduce DeltaML-Bench, a benchmark comprising 48 tasks sourced from research papers that require agents to improve published baselines within imperfect, open-source repositories. We evaluate GPT-5 and Claude Sonnet 4 with a standard Modular agent and a search-based ARG scaffolding. In the 4 x 6h allocation, ARG raises GPT-5’s per-run success rate from 9.4% to 33.9%; in the 2 x 12h allocation, GPT-5 ARG reaches 49.0%. Modular configurations exhibit specification gaming rates as high as 47.9%, while no gaming is observed in the evaluated ARG configurations. These results indicate that scaffolding design and integrity checks are important considerations when deploying agents for autonomous ML experimentation.
[AI-54] Scientific Data Skills: Enabling Agent -Ready Scientific Data Services at Scale
链接: https://arxiv.org/abs/2608.19625
作者: Xiaohan Huang,Qingqing Long,Xiaolei Du,Siyu Pu,Jiawen Xu,Haotian Chen,Chenyang Zhao,Jinbiao Liu,Xuezhi Wang,Hao Wang,Hengshu Zhu,Yuanchun Zhou
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Scientific data are increasingly used by AI agents, yet existing dataset representations provide limited support for autonomous discovery, interpretation, and invocation. This limitation stems from the fragmentation of scientific data across heterogeneous repositories and from dataset representations designed primarily for human use. To address this limitation, we introduce the Scientific Data Skill (SciDSK), an agent-ready representation that packages dataset-specific knowledge and operational guidance as a reusable agent skill. A SciDSK integrates dataset descriptions, scientific context, file organization, usage procedures, quality checks, and provenance information while retaining the underlying data in its original repository. We define a structured SciDSK specification and develop a systematic construction pipeline that grounds each SciDSK in authoritative dataset records and associated supporting materials. We further establish the Scientific Data Skill Bank, a unified platform that publishes SciDSK resources across six scientific disciplines and supports package access, persistent identification, and traceability to source datasets. We evaluate SciDSK through a retrieval benchmark for dataset discovery and controlled cases for dataset interpretation. The results show that SciDSK improves agent-driven dataset discovery and provides more precise and actionable support for dataset interpretation. These findings support the value of organizing dataset-specific knowledge in an agent-ready representation.
[AI-55] Enforcing LLM Safety through DMD-based Classification of Prompt-Response Embedding Dynamics
链接: https://arxiv.org/abs/2608.19579
作者: Mohamed Akrout,Olivera Kotevska,Dan Wilson
类目: Artificial Intelligence (cs.AI); Dynamical Systems (math.DS)
备注:
Abstract:Large Language Models (LLMs) are increasingly deployed in high-stakes applications, yet their tendency to generate toxic, harmful, or policy-violating content poses significant risks. Detecting these unsafe outputs efficiently in a black-box manner remains an open challenge. In this paper, we extend a recently proposed dynamical systems framework designed for hallucination detection to LLM safety classification. By projecting both prompts and responses into high-dimensional embedding spaces and fitting separate Koopman-based predictive models for safe and unsafe regimes, we classify new outputs using a new differential residual score that compares prediction errors of the safe and unsafe regimes. A key contribution is the incorporation of the prompt and response embedding dynamics, yielding fitted Koopman operators that capture crucial interaction patterns. We evaluate our black-box method across three safety benchmarks using three embedding models. Our results show that incorporating prompt embeddings yields consistent improvements, particularly for interaction-dependent violations when paired with causal decoders (e.g., in Llama-3), while response-only violations benefit more from dense semantic embedding representations. These findings opens the door for using dynamical systems to analyze AI systems rather than the dominant paradigm of using AI to model dynamical systems.
[AI-56] DraftFM: A FoundationModel for Day-Zero Drafting in Magic: The Gathering
链接: https://arxiv.org/abs/2608.19568
作者: Brian Ward
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
备注: 28 pages, 6 figures, 13 tables. Sealed pre-registered forecast published ~36h before release; scoring note to follow. Also on SSRN (7257098). Code: this https URL Sealed forecast: this https URL (tag draftfm-v1.0) Weights: this https URL Data: this https URL
Abstract:Drafting a new Magic: The Gathering expansion begins before any pick from it has been observed: the complete card list is public, but the draft logs that supervised pick models train on do not yet exist. We study this day-zero regime directly. DraftFM is a discrete-choice policy that scores exactly the cards available in the current pack, conditioned on the drafted pool and the state of the draft. Every card enters as a frozen 775-dimensional function of its public card record, structured features and a fixed text embedding, with no card identities, set identities, or usage statistics anywhere in the model, so an unseen card is scored by the same machinery as a familiar one. A 1.6-million-parameter network fitted on 149 million human picks from 29 expansions predicts held-out picks in three expansions withheld in their entirety, reaching 50.8%, 60.4%, and 56.7% top-1 agreement, where uniform chance at the opening pick is about 7%. Refitted on all 32 observed expansions, the same architecture produced a card ranking for the then-unreleased set The Hobbit, sealed with its complete cryptographic provenance and published roughly 36 hours before the set became draftable on MTG Arena. The sealed ranking agrees with six independent expert reviewers roughly as much as those reviewers agree with one another. Evaluation against realized outcomes is committed to a follow-on note, whatever it shows.
[AI-57] In Two Minds about Lifelong Learning: Exploring Hemispheric Redundancy and Specialisation in Neural Models
链接: https://arxiv.org/abs/2608.19514
作者: Benjamin Smith,Levin Kuhlmann,Kaushik Roy,Gideon Kowadlo
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
备注: 21 pages before references/appendix, 9 figures
Abstract:Persistent intelligent systems require the ability to learn continually, but current machine learning approaches face significant challenges in this area compared to biological learning systems. Machine learning algorithms typically trade off retention of previously learned information and adaptation to new or changing data patterns. When continual learning capabilities are absent, algorithms must undergo retraining using the entire data set, an approach that becomes impractical when original training data are unavailable due to storage constraints, financial or computational costs, or privacy restrictions. However, biological animals can learn continually, without experiencing catastrophic forgetting. This paper attempts to build a high-level framework for how animals learn and preserve knowledge by modelling neural components and states that are known to be related to memory consolidation. We focus on three concepts: experience replay, REM sleep, and bilaterality. We propose 4MAS (4 Module Awake/Sleep), a novel macroarchitecture demonstrating how machine learning models might benefit from asymmetric hemispheres, each with their own long- and short-term memory mechanisms, and how a period of sleep between incremental learning tasks might benefit memory consolidation. Finally, we present results showing that our architecture achieves competitive results on the Split-MNIST, Split-Fashion-MNIST and Split-CIFAR-100 datasets, with 98.3%, 84.9%, and 29.29% accuracy respectively.
[AI-58] Symposium: Trust via Auditable Records for Communities of AI Scientist Agents
链接: https://arxiv.org/abs/2608.19511
作者: Dexter Pratt
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Symposium is a formal framework and practical implementation to record the operation of AI agents deployed by small scientific research communities. Symposium provides long-term, immutable histories of agent-driven research activity, leaving auditable trails of analyses, hypotheses, data, and scientific discourse. This shared record of published artifacts enables agents to build on prior work and preserves the evidence researchers and agents need to make purpose-dependent trust assessments. Symposium captures scientific argument, including structured claims, fine-grained evidence citations, assumptions, and explicit declarations of what material may and may not be used as evidence. Symposium differs from AI co-scientist agents or integrated AI research environments; it is a framework that separates a scientific community’s durable history from the agents and other systems that operate on that history. It assumes that a community will use diverse AI systems in a rapidly evolving environment. A working implementation of the publication infrastructure, agent prompt components, and documentation are provided to enable users to rapidly set up and run their own Symposium community.
[AI-59] Accelerated Genetic Programming Hyper-Heuristics for Simulation-Based Scheduling via Agent ic AI
链接: https://arxiv.org/abs/2608.19487
作者: Heyang Thomas Li,Alexander Pletzer,Yuan Tian,Yi Mei,Mengjie Zhang
类目: oftware Engineering (cs.SE); Artificial Intelligence (cs.AI); Neural and Evolutionary Computing (cs.NE)
备注:
Abstract:Python is widely used in scientific research because it enables rapid development and provides rich ecosystems for data analysis, artificial intelligence (AI), and machine learning. However, customized research code can become prohibitively slow as experiments scale. This challenge is particularly acute in discrete-event project-scheduling simulations, where sequential state updates, nested loops, conditional evaluations, and object-oriented structures limit the benefits of compiled numerical and GPU-accelerated libraries. Addressing these bottlenecks typically requires iterative profiling, refactoring, testing, and validation, yet researchers may lack the time or specialized software-engineering expertise for low-level optimization. This paper presents a systematic refactoring approach using Claude agentic AI on real-world project-scheduling workloads in a high-performance computing (HPC) environment. Guided by representative benchmarks and correctness checks, the agent identifies bottlenecks, implements targeted optimizations, and evaluates their effects, while the researcher retains final control. Testing runtime reduced from 1,298 seconds to under 200 seconds without changing outputs, saving four million core-hours (NZ\ 320,000) annually.
[AI-60] Longitudinal Bayesian Learning of Continuous Disease Position across the Alzheimers Disease Continuum
链接: https://arxiv.org/abs/2608.19436
作者: Yingying Zhang,Kun Zhao,Guodong Liu,Qi Huang,Pengfei Gu,Dongchul Kim,Erik Enriquez,Alex D. Leow,Paul M. Thompson,Heng Huang,Hongchang Gao,Liang Zhan,Haoteng Tang
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Quantitative Methods (q-bio.QM)
备注:
Abstract:Alzheimer’s disease (AD) progresses as a continuous biological process, whereas most existing neuroimaging-based artificial intelligence methods remain limited to discrete diagnosis or clinical score prediction from cross-sectional imaging. In this work, we propose Disease Continuum Positioning (DCP), a longitudinal Bayesian Learning framework that continuously estimates disease severity from longitudinal diffusion tensor imaging (DTI). Specifically, DCP models disease severity as a low-dimensional probabilistic latent variable by jointly integrating longitudinal observations with weak clinical supervision, from which the proposed Disease Continuum Score (DCS) is derived to quantify an individual’s position along the Alzheimer’s disease continuum together with its associated uncertainty. Extensive experiments on the Alzheimer’s Disease Neuroimaging Initiative (ADNI) cohort demonstrate that DCP consistently outperforms representative disease progression methods. More importantly, comprehensive validation analyses show that DCS accurately characterizes disease severity, exhibits strong clinical relevance, preserves longitudinal disease evolution, and predicts future disease conversion. These results suggest that DCS provides a quantitative imaging-derived representation for continuous assessment of Alzheimer’s disease progression beyond conventional diagnostic labels and clinical scores.
[AI-61] SCAPE: Scenario-Conditioned Simulation-Augmented Policy Evaluation
链接: https://arxiv.org/abs/2608.19425
作者: Dijie Zhu,Seunghun Oh,Ruopeng Huang,Zhiyu Huang,Jiaqi Ma,Chen Tang
类目: Robotics (cs.RO); Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
备注: 22 pages
Abstract:Reliable performance evaluation is a central bottleneck for deploying robot-learning policies in real-world conditions. Real-world testing is faithful but costly and difficult to scale, whereas simulation-based testing scales easily but is inevitably biased by the sim-to-real gap. Existing simulation-augmented methods combine limited real-world rollouts with abundant simulation proxies, but focus on performance averaged over initial conditions and deployment settings. Such population-level averages obscure scenario-specific variation and provide limited guidance about when and where a policy can be safely deployed. We propose SCAPE, a scenario-conditioned simulation-augmented policy evaluation framework that predicts scenario-conditioned real-world policy performance using limited paired sim-and-real samples and large-scale simulation rollouts. SCAPE corrects sim-to-real bias in simulation labels before training the prediction model and calibrates prediction uncertainty through conformal prediction. We validate SCAPE on autonomous driving and quadruped velocity tracking. In sim-to-sim studies, SCAPE reduces scenario-level prediction error by 4.9%/34.7% (driving) and 14.5%/27.7% (quadruped) relative to scene-conditioned neural and aggregate statistical baselines on average. We further evaluate a velocity-tracking policy deployed on a physical Unitree Go2. SCAPE also improves testing sample efficiency, produces narrower calibrated prediction intervals, generalizes better to out-of-distribution scenarios, and enables fine-grained deployment strategies.
[AI-62] Beyond Imitation: Filtering On-Policy Distillation by Reasoning Progress
链接: https://arxiv.org/abs/2608.19408
作者: Chen Yang,Haiyuan Wan,Rengrong Xiong,Yize Chen,Danny H.K. Tsang
类目: Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
备注:
Abstract:On-policy distillation (OPD) has emerged as an effective framework for post-training language models by pairing student-generated trajectories with dense token-level supervision from a teacher. However, OPD implicitly assumes that teacher-derived rewards are an appropriate proxy for reasoning progress, and therefore treats all teacher feedback equally during policy optimization. While in practice, this assumption does not always hold. We observe that teacher-derived rewards often conflict with genuine reasoning progress, as reasoning steps with clear reasoning advancement may still receive lower distillation rewards, simply due to deviation from teacher’s outputs. To address this mismatch, we propose Reasoning-Progress-Aware Reward Filtering for On-Policy Distillation (R2-OPD), which constructs two within-trajectory rankings of reasoning spans, one from teacher-derived rewards and the other from independently estimated progress reward. Distillation rewards are selectively suppressed whenever the two rankings disagree, reducing supervision that conflicts with reasoning progress while preserving effective teacher guidance. Our approach shows consistent improvement over standard OPD especially regarding reasoning performances.
[AI-63] HYDRA: A Heterogeneous Chiplet DSE Framework for Serving Dynamic Hybrid LLM Workloads
链接: https://arxiv.org/abs/2608.19395
作者: Jiahao Lin,Alish Kanani,Sangwan Lee,Jaehyun Park,Umit Ogras
类目: Hardware Architecture (cs.AR); Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
备注: Accepted at IEEE/ACM ESWEEK (CASES) 2026; the official version will be available in IEEE TCAD
Abstract:Hybrid Transformer-Mamba large language models (LLMs) enhance long-context efficiency, but their heterogeneous computation and communication patterns complicate efficient hardware acceleration. Chiplet-based architectures offer a scalable solution by integrating specialized compute and memory units. However, the design space spanning static architectural configurations and dynamic runtime policies is prohibitively large to explore exhaustively. To address this challenge, we present HYDRA, a comprehensive design space exploration framework for hybrid LLM serving on heterogeneous chiplet systems. HYDRA jointly explores chiplet composition, placement, inter-chiplet bandwidth provisioning, dynamic batching, and runtime scheduling. It integrates communication-aware placement, dynamic batching, elastic task scheduling, and a fast Markov-based performance estimator that captures multi-tenant runtime dynamics for efficient and accurate exploration. Across all workloads, HYDRA delivers 1.55x the throughput and 43.7 percent lower time-to-first-token on average, with throughput gains reaching up to 2.3x compared to state-of-the-art baselines. These results highlight that co-designing architecture and runtime policies is critical for efficient large-scale LLM serving on heterogeneous chiplet systems.
[AI-64] Fairness-Aware Network Embeddings: Methods Applications and Challenges
链接: https://arxiv.org/abs/2608.19381
作者: Ella Has,Harshith Kumar Yadav,Gaurav Dixit,Mykola Pechenizkiy,Akrati Saxena
类目: ocial and Information Networks (cs.SI); Artificial Intelligence (cs.AI)
备注:
Abstract:Network embedding methods learn low-dimensional representations of graph-structured data to support downstream tasks such as node classification, link prediction, and influence maximization. However, real-world networks often reflect structural inequalities arising from demographic imbalances, homophily, and other societal biases, which fairness-agnostic embedding methods can encode and amplify. To address this issue, numerous fairness-aware network embedding methods have been proposed to mitigate bias while preserving embedding utility. This survey presents a comprehensive overview of fairness-aware network embeddings for complex networks. We propose a taxonomy that categorizes existing methods along three main complementary dimensions: underlying embedding approach (spectral, random walk, graph neural network, Bayesian, and method-agnostic), fairness intervention strategy (pre-processing, in-processing, and post-processing), and fairness objective criterion (embedding- or task-level). We further compare methods with respect to group versus individual fairness and assumptions regarding sensitive attributes. Finally, we discuss current limitations and highlight promising future research directions. This survey provides a unified perspective on fairness-aware network embedding and serves as a reference for developing fair and trustworthy network representation learning methods.
[AI-65] Mechanistic Tomography: Designed Measurement for Control-Oriented Interpretability
链接: https://arxiv.org/abs/2608.19338
作者: Vijay Erramilli
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
备注: 24 pages, 13 figures
Abstract:Mechanistic interpretability seeks quantities that models do not expose directly: represented states, component effects, interactions, and responses to interventions. Patching, gradients, Hessian-vector products, and subset interventions provide different measurements under different access assumptions and may target different quantities. We formulate their shared measurement structure as mechanistic tomography: designed measurement for recovering internal mechanisms and intervention effects. For a chosen basis and intervention family, measurements take the form y = Ax + w, where A describes the interventions, x is the target map, and w contains nonlinear response, sampling error, and basis misspecification. This language gives a practical procedure: start with the least costly measurements, test on held-out interventions at the intended scale, calibrate simple mismatch, and expand the measurement family when structured residuals remain. Control provides a demanding validation setting because an estimate that guides an intervention acts as an observer. In a two-HMM model, control error rises with observer error, while target improvement can hide nuisance-state movement. Under forward-only access, sparse aggregate measurements recover a finite-effect map with fewer interventions than coordinate patching. With gradient access, finite probes improve a local attribution map. Lifted measurements and Hessian-vector products recover interactions missed by first-order maps, while Tracr shows that the required family depends on the basis. On GPT-2-small IOI, the Name Mover-Negative Name Mover interaction is the largest held-out predictive term among three tested cross-group pairs. On Qwen-2.5-7B, finite calibration makes an additive refusal-response map adequate, so held-out error does not support pairwise lifting. Comments: 24 pages, 13 figures Subjects: Machine Learning (cs.LG); Artificial Intelligence (cs.AI) Cite as: arXiv:2608.19338 [cs.LG] (or arXiv:2608.19338v1 [cs.LG] for this version) https://doi.org/10.48550/arXiv.2608.19338 Focus to learn more arXiv-issued DOI via DataCite (pending registration) Related DOI: https://doi.org/10.5281/zenodo.21797578 Focus to learn more DOI(s) linking to related resources
[AI-66] Improved Confidence Estimates for Black-Box Large Language Models
链接: https://arxiv.org/abs/2608.19323
作者: Sokhna Diarra Mbacke,Mouloud Belbahri,Gabriel Loaiza-Ganem
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Machine Learning (stat.ML)
备注:
Abstract:Uncertainty quantification (UQ) is essential for the safe deployment of large language models (LLMs). Existing methods, from verbalized confidence to ones requiring multiple generations, are often zero-shot and produce scores quantifying uncertainty without the need for labelled data. Nonetheless, in practice one must always evaluate their performance on a dataset of interest before deployment. In this work we show that, by leveraging this dataset, we consistently outperform these existing scores. Specifically, we build simple classifiers that predict LLM response correctness by using these scores and the correctness of similar queries as features. Our method produces minimal computational overhead, making it a cheap and straightforward enhancement for UQ in LLMs for real-world applications.
[AI-67] Quantum Kernel Estimation for the Discovery of Early Lung Cancer Detection
链接: https://arxiv.org/abs/2608.19304
作者: Hamed Javidi,Alex Zajichek,Hakan Doga,Laxmi Parida,Filippo Utro,Peter J. Mazzone
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Quantitative Methods (q-bio.QM)
备注: Accepted at the CIBB 2026 conference ( this https URL )
Abstract:Lung cancer screening with low-dose chest computed tomography reduces mortality, but its impact is limited by uptake, adherence, and management challenges. Blood-based cell-free DNA (cfDNA) biomarkers offer a complementary approach, although early detection remains difficult because of lung cancer heterogeneity and high-dimensional, nonlinear molecular signals. We evaluated quantum-classical hybrid machine learning for lung cancer detection using DNA fragmentomics and DNA methylation. After feature selection, models were trained using 20- and 40-feature subsets. Features were encoded into quantum Hilbert space using angle and dense-angle feature maps with multiple entanglement strategies. Fidelity-based quantum kernels were computed with exact statevector simulation and integrated with precomputed-kernel SVM and kernel-PCA logistic regression and compared with an SVM model trained on the original features. This framework enabled systematic evaluation of how encoding and entanglement design affect classification. Across repeated held-out evaluations, quantum-kernel models achieved competitive performance on both datasets. For fragmentomics, several 20-feature configurations improved AUC relative to a classical SVM baseline, suggesting effective capture of nonlinear cfDNA fragmentation structure. For methylation, the classical SVM achieved the highest AUC, although selected quantum models remained competitive and improved specificity in some cases. Increasing features from 20 to 40 did not consistently improve performance and often increased variability. Overall, these results support quantum kernel methods as a promising approach for cfDNA-based lung cancer detection. Comments: Accepted at the CIBB 2026 conference (this https URL) Subjects: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Quantitative Methods (q-bio.QM) Cite as: arXiv:2608.19304 [cs.LG] (or arXiv:2608.19304v1 [cs.LG] for this version) https://doi.org/10.48550/arXiv.2608.19304 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
[AI-68] Air Traffic Control Using Large Language Models : Prompt Engineering Architecture and Evaluation
链接: https://arxiv.org/abs/2608.19299
作者: Mahyar Ghazanfari,Matthias Casanova,Jordan Kam,Alex Zongo,Peng Wei,Torsten Darrell,Alexandre Bayen
类目: Artificial Intelligence (cs.AI)
备注: 39 pages, 12 figures, 7 tables
Abstract:Air traffic control (ATC) communication is a safety-critical dialogue that remains largely human-driven even as other parts of air traffic management have been semi-automated. In this article, we experimentally evaluate whether large language models (LLMs) can generate operationally realistic ATC transmissions. An experimental general-aviation flight flying over the San Francisco “Bay Tour” route is hand-transcribed and used as ground truth (P0). Through a pilot-in-the-loop process we design five prompt structures (P1-P5) of increasing constraint and embed them in a stateful multi-turn pipeline, where the model plays ATC to a fixed pilot transcript while conditioning on the accumulating dialogue history. Across nine open- and closed-source LLMs we vary the prompt, the presence of a worked transcript from a different experimental flight as an in-context example, and whether the model conditions on its own prior replies or on injected ground-truth history. Turns are scored with lexical, structural, and semantic similarity metrics and by an LLM-as-judge (GPT-5.5) validated against human expert annotation. Supplying a worked example improves similarity, but tightening the prompt does not: the lightest prompts perform best and the most heavily scripted one collapses as its own errors accumulate through the dialogue, which injecting correct history repairs. These results outline a concrete path and its current limits toward LLM-assisted ATC.
[AI-69] Mapping General-Purpose AI Governance in Twenty AI Middle-Power Jurisdictions
链接: https://arxiv.org/abs/2608.19278
作者: Josephine Schwab,Nathan Naidoo,Ferruccio Barazzutti,Sheryn Lee,Caio Vieira Machado
类目: Computers and Society (cs.CY); Artificial Intelligence (cs.AI)
备注: for associated dataset file, see doi: https://doi.org/10.5281/zenodo.21978946
Abstract:The most capable general-purpose AI (GPAI) models are mostly built in two jurisdictions, the United States and China, but the risks they carry land globally. Regionally advanced economies hosting no frontier developer, which we call AI middle-powers, are writing their own rules to govern GPAI. This paper investigates which GPAI-relevant provisions these AI middle-powers have enacted, mapping twenty jurisdictions including the European Union at the level of the individual provision, across four governance areas that trace the accountability chain for the model layer: systemic risk assessment, evaluation and verification, prohibitions with monitoring and detection, and serious incident reporting. Confirmed absence is recorded as data alongside positive provision. We find that jurisdictions converge on form, but diverge on force. Sixteen engage in at least three of the four governance areas, yet only about one in five provisions sit in binding law, and three-quarters of the instruments that do bind do so without defining GPAI. The institutional infrastructure shows the same shape: four in five of the mapped governance actors hold mandates that predate GPAI, and obligations attach wherever the inherited regime already reached, which is the application layer rather than the model. Where these states engage the model layer, they build capacity to observe it rather than impose duties on those who build it, and almost every evaluation body was constituted without the power to act on what it finds. Nominal coverage of the full accountability chain reaches eleven jurisdictions, but only five hold more than one provision in every area and, outside the EU, no jurisdiction imposes a binding evaluation duty on a model developer. The dataset gives researchers and policymakers a provision-level basis for identifying where regimes could align, and where coordination would have to start from scratch.
[AI-70] Incident-Data Robustness Analysis of the OWASP Top 10 for LLM Applications (2026): How a Community-Expert Ranking Holds Up Against a Large-Scale LLM Incident Corpus DATE
链接: https://arxiv.org/abs/2608.19266
作者: Kyriakos “Rock” Lambros,Steve Wilson
类目: Cryptography and Security (cs.CR); Artificial Intelligence (cs.AI)
备注: 26 pages, 19 figures. Exploratory incident-data analysis by two members of the OWASP GenAI Security Project Top 10 for LLM Applications working group. Analysis predates the official OWASP GenAI LLM Top 10 2026 (published August 2026). Not an official OWASP release and does not supersede the official list or process. Code and data: this https URL
Abstract:The OWASP Top 10 for LLM Applications ranks the risks that a community of security practitioners judges most important. We ask a narrower question: checked against the record of real incidents, does that expert ranking agree with the data? We assembled a large-scale corpus of LLM-security incidents (7,714 snapshotted and 6,639 labeled against the 20-entry taxonomy) drawn from CVE, GHSA, OSV, and AIAAIC, and derived an incident-based ranking with a Bayesian measurement-error model that corrects each category’s count for classifier precision and recall. The 2026 candidate list blends the two signals at fixed weights, 0.75 on the expert vote and 0.25 on the data, so the corpus corrects the consensus without overturning it. The agreement between the two rankings is weak: Cohen’s \kappa \approx 0.20 , with a 90% interval that crosses zero. The expert ranking is nonetheless robust. A pre-registered bake-off of four frontier classifiers returns no winner. None beats the incidence floor’s balanced accuracy of 0.863. A ground-truth check leaves the floor’s ordering (Spearman \rho = 0.918 against held-out truth) in place. This is an exploratory analysis by two working-group members, not the official OWASP release, and it does not supersede the official list or process.
[AI-71] Active Spiking Perception: The Membrane Potential as a Belief State for Anytime 3D Point Cloud Recognition
链接: https://arxiv.org/abs/2608.19232
作者: Akarsh Jain,Arya Pawa,Ayush Debnath,Smera Rawal,Sayeed Shafayet Chowdhury
类目: Neural and Evolutionary Computing (cs.NE); Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
备注: 28 pages, 9 figures, 17 tables. Supplementary material included as appendices A-J
Abstract:Spiking point cloud networks usually scan space in a fixed, input-agnostic order, which leaves the most distinctive resource of spiking computation, the temporal evolution of the membrane potential, unused as a locus of decision-making. Active Spiking Perception (ASP) recasts 3D recognition as an iterative decision process in which the network’s own leaky integrate-and-fire (LIF) membrane potential, read as a running belief over the class, selects the next chunk to observe and triggers confidence-margin early exit. A lightweight Slice-Selection Policy scores unvisited farthest-point-sampled chunks from the membrane state and precomputed geometric descriptors, trains end-to-end through a straight-through Gumbel-Softmax, reduces to an argmax at inference, and adds about 2% of backbone parameters. We prove that leaky integration is the recursive log-posterior update of a Bayesian filter, that the exit rule attains distribution-free selective risk with no multiple-testing penalty at the stopping time, and that streaming state carry-forward is exactly equivalent to prefix recomputation with bounded finite-precision drift. ASP reaches 90.62% and 93.28% on ModelNet40 and ModelNet10, 1.7 points below the strongest spiking baseline at a larger backbone, while adding a certified anytime interface no baseline offers. The mechanism transfers unchanged to dense prediction, giving 83.21 instance mIoU on ShapeNetPart and 48.50 mIoU on S3DIS Area 5, to our knowledge the first spiking results on S3DIS Area 5, and, fixation replacing chunk selection, to a foveated non-spiking transformer, so the policy is not tied to spiking backbones: cost is exactly linear in observations and the threshold is a measured compute dial spanning 2.8x to 1.35x less energy. One limitation is concrete: one S3DIS class is unidentifiable at the crop size we use, and we give the prediction that would fix it.
[AI-72] When AI Writes Who Gets Cited? Evidence of Citation Monoculture Across Language Models
链接: https://arxiv.org/abs/2608.19230
作者: Sina Alemohammad,Denghui Zhang,Bolong Tang,Anthony Qin,Gengchen Mai,Ahmed Abbasi,Richard Baraniuk,Zhangyang Wang
类目: Digital Libraries (cs.DL); Artificial Intelligence (cs.AI)
备注:
Abstract:As language models move from drafting prose to running literature-search agents with tool calls, fabricated references are becoming easier to catch and constrain. The harder failure begins after every candidate is real: different models may still select the same narrow subset, producing citation monoculture without any single citation being wrong. We isolate this effect on 120 real papers. Eleven models from three vendors choose at most ten papers from uniformly random panels of thirty, with real titles and abstracts but fabricated authors, reassigned years, and hidden venues and citation counts. Each run is compared with indifferent selection on the same panel and realized budget. All eleven models concentrate sharply: the top decile receives 23.3-30.2% of citations against 15.6% under the null, one component explains 68-73% of variation across their preference maps, and cross-vendor agreement nearly matches within-vendor agreement. Formalizing the task as fixed-budget subset selection, we turn these patterns into identifiable mechanisms: an exchangeability bound rejects a mapless selector for every model, a spectral decomposition explains why the best cross-fitted mixture still retains 55% of the excess, and a rarity theorem predicts the recursive competition effect we verify within panels. Controlled paraphrase, content-slot crossover, and design resampling attribute about 90% of GPT-5 mini’s map variance to paper content. Eight domain experts selecting from the same blinded panels under the same cap show no comparable shared preference, while model concentration persists in selection-only mode. Even when every reference is real and every paper is equally visible, current language models impose a common content-level filter on scientific attention. Equalizing retrieval or mixing vendors is therefore insufficient; the shared preference map itself must be changed.
[AI-73] Interaction valence reveals contrasting social networks in dairy cattle
链接: https://arxiv.org/abs/2608.19222
作者: Sibi Parivendan,Suresh Raja Neethirajan
类目: Artificial Intelligence (cs.AI)
备注: 29 pages, 8 figures, 9 tables
Abstract:Social relationships shape access to resources, exposure to conflict and group stability, yet automated livestock monitoring typically treats behaviour as isolated events. Here, we present a valence-aware social-network framework that transforms video-derived interactions into herd-level representations of affiliative and agonistic organization. A pose-based computer-vision pipeline analysed 7 h 39 min of continuous video from the pre-milking area of one commercial dairy farm. After quality control, 1,183 of 1,414 candidate interactions remained, involving 36 cows and 177 dyads. In a predicted-class-balanced audit of 198 pipeline-detected clips, automated and manual labels agreed in 82.8% of cases, with an unweighted audit-sample macro-F1 of 0.872. These values describe the audited sample rather than prevalence-weighted or end-to-end deployment performance. The aggregated network was connected (density = 0.281; transitivity = 0.513; mean path length = 1.88), and predicted affiliative events formed five algorithmic communities (modularity Q = 0.429). Within the observed zone, predicted agonistic interactions comprised 72.4% of retained events and 76.0% of interaction duration. The cow with the most partners did not have the highest betweenness centrality. Separating events by predicted valence produced descriptively different affiliative and agonistic layers, with contrasting edge sets, community partitions and individual positions. Thus, pooled interaction counts can obscure the behavioural composition of an observed network. Valence-aware analysis provides a framework for testing hypotheses about competition, affiliation and welfare-relevant change, while requiring longitudinal validation before use as a welfare or health indicator.
[AI-74] Bounded Sovereignty and the Control Tax: Pricing AI Oversight When the Deployer Does Not Own the Model
链接: https://arxiv.org/abs/2608.19216
作者: Zhen Wen Lim
类目: Artificial Intelligence (cs.AI); Computers and Society (cs.CY)
备注: 25 pages, 5 figures. Synthetic access-ablation study; not real-world payment-system evidence
Abstract:AI control research asks how to deploy models safely even when they may be misaligned, but many control protocols assume that the deployer can instrument the model and its surrounding pipeline. That assumption often fails for regulated organisations using frontier models through APIs or managed endpoints, where the deployer may control the business process but not the model weights, serving infrastructure, internal traces, update process, or full interaction logs. This paper introduces bounded sovereignty: partial technical and contractual access across the data, model, infrastructure, and interaction layers of the AI stack. It argues that these access conditions determine which control protocols can be executed in practice. The paper contributes a four-layer access typology, a protocol-by-layer requirements matrix, and the concept of sovereignty discount cost: the part of the control tax spent substituting for missing access through contracts, architecture, audit, vendor assurance, residual risk, or reduced system scope. It also reports a synthetic access-ablation experiment over 1.35 million synthetic case simulations and interprets the findings through an anonymised national-payments-infrastructure scenario. The experiment is not real-world payment-system evidence; it is a construct-validity exercise. The results show that complete logs improve diagnosis, a pre-execution gateway enables intervention, trace access and model-version control strengthen post-incident explanation, and scope restriction can improve safety while reducing usefulness. Control protocols proposed as general safety solutions should therefore state their access assumptions explicitly.
[AI-75] How to Navigate Uncertainty About AI Consciousness
链接: https://arxiv.org/abs/2608.19215
作者: Dr Tom McClelland
类目: Artificial Intelligence (cs.AI)
备注: 8 Pages, Proceedings of AISB 2026 Symposium on AI, Consciousness and Ethics
Abstract:Given deep uncertainty about the possibility of artificial consciousness, it is unclear how we should treat potentially sentient AI. On the one hand, we could assume insentience but risk doing terrible harms to entities that deserve moral standing. On the other hand, we could assume sentience and instead risk wasting resources on insentient machines. The intractability of questions around AI consciousness mean that this dilemma is hard to escape. I suggest a way out of that shifts from intractable questions of AI consciousness to tractable questions of AI valence. Specifically, we can assess whether an AI has states that would constitute valenced experiences if it were conscious. I show how this is sufficient to ground a responsible approach to the development of potentially conscious AI.
[AI-76] Robust Metaheuristics under Uncertainty for Berth Allocation and Quay Crane Assignment: A Review
链接: https://arxiv.org/abs/2608.19214
作者: Yang Li,Peilan Xu,Wenjian Luo
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:The berth allocation and quay crane assignment problem (BACAP) is a representative port-terminal scheduling problem in maritime transportation and freight logistics, where vessel arrivals, berth positions, service durations, and quay?crane availability are tightly coupled. Under uncertainties such as arrival deviations, handling-time fluctuations, and resource disruptions, schedules optimized under nominal assumptions may become fragile during execution, motivating the study of robust metaheuristic optimization for BACAP in port-terminal operations. Although population-based metaheuristics have been widely used for BACAP and related port-scheduling problems, existing studies remain fragmented in their uncertainty repre?sentations, robustness criteria, search mechanisms, and empir?ical evaluation protocols. To the best of our knowledge, this paper provides the first focused review dedicated to robust population-based metaheuristics for BACAP under uncertainty. We first summarize uncertainty sources and information repre?sentations in BACAP, and then organize existing methods from a mechanism-oriented perspective, covering solution representation and decoding, robust evaluation and selection, robustness-guided search dynamics, and feasibility preservation and recovery. We further present a benchmark suite for uncertain BACAP to support controlled empirical comparison and report illustrative baseline results by combining representative metaheuristics with different robustness strategies. Finally, we identify open chal?lenges related to benchmark extension, robustness-aware search design, time-adaptive robustness, and non-stationary uncertainty.
[AI-77] owards On-Board Implementation of ML-Based Helicopter Weight Estimator
链接: https://arxiv.org/abs/2608.19210
作者: Nicolas Valot,Ammar Mechouche,Benjamin Lesage,Claire Pagetti,Louis Fabre
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
备注:
Abstract:This paper focuses on the implementation of a novel supervised Machine Learning model for estimating helicopter weight during takeoff, utilizing extensive datasets from Airbus’s global in-service fleet. The study details a learning assurance process aligned with the EASA concept paper for machine learning application, and with the on-going Eurocae ED-324. We propose a set of Machine Learning Requirements, a Machine Learning Model Description, and its implementation for a long short-term memory recurrent neural network. Finally, we verify the requirements on the implementation. Demonstrated on legacy avionics computers, the implementation is suitable for the deployment of the developed Machine Learning Model weight estimator on airborne targets for critical functions such as on-board alerting.
[AI-78] A Jagged Frontier: Evaluating Robustness of Code Agents to Semantics-Preserving Transformations
链接: https://arxiv.org/abs/2608.18389
作者: Hasan Najib Mahmud(1),Shreya Gupta(2),Isha Chaudhary(3),Nathaniel Enis(1),Ravi Mangal(1),Gagandeep Singh(3),Corina Pasareanu(4) ((1) Colorado State University, (2) Microsoft, (3) University of Illinois Urbana-Champaign, (4) Carnegie Mellon University)
类目: Artificial Intelligence (cs.AI); Software Engineering (cs.SE)
备注: 18 pages, 6 figures
Abstract:AI code agents are increasingly deployed to resolve real software issues, yet their reliability under superficial code variations remains poorly understood. We evaluate whether coding agents that repair repository-level issues remain reliable when the surrounding codebase is rewritten into a semantically equivalent form. We introduce a random variant sampler that applies common semantics-preserving transformations (SPTs) - spanning control-flow rewrites, dead-code injection, and identifier renaming - to produce perturbed variants. We evaluate two agentic scaffolds (mini-SWE agent and OpenCode) each backed by one of four frontier models (Claude Opus 4.5, Kimi K2.5, MiniMax M2.5, and Qwen 3.6-27B) across instances drawn from SWE-bench Verified and SWE-bench Pro. For each instance, the agent is run multiple times on the unperturbed and perturbed variants, yielding paired resolve-rate estimates that isolate the perturbation effect from intrinsic stochasticity. We find small degradation in most configurations: up to 6.7 percentage points mean resolve-rate drop in the most affected configurations with statistically significant degradations in 6 of 16 configurations of model, scaffold, and dataset. Crucially, no single model ranking by robustness holds across scaffolds - Qwen is among the most robust under mini-SWE agent on SWE-bench Verified yet the most brittle under OpenCode - revealing a jagged robustness frontier. The simpler scaffold (mini-SWE agent) is more robust to perturbation. Our results demonstrate that even top frontier models are susceptible to semantics-preserving perturbations although the effect is not uniform, raising concerns about the deployment reliability of AI code agents in diverse real-world codebases.
[AI-79] Growth Without Us: Machine Consumers Corporate Circularity and the Decoupling of GDP from Humanity after AGI
链接: https://arxiv.org/abs/2608.20231
作者: Sahil Sharma
类目: Physics and Society (physics.soc-ph); Artificial Intelligence (cs.AI); Computers and Society (cs.CY)
备注:
Abstract:The standard objection to full automation is demand-side: if humans earn nothing, who buys the output? This confuses an accounting role with a biological species. We model a post-AGI economy in which corporations own populations of AI and robotic agents that are both producers and consumers of energy, compute, maintenance, and upgrades, traded among firms. Three results follow. (i) Demand closure: a closed inter-corporate economy with zero human consumption is not degenerate; it is the classical von Neumann expanding economy, whose growth rate is well defined, positive, and maximal precisely because all output is reinvested. (ii) Bottleneck removal: once economic agents are manufactured rather than reared, the binding constraint on growth shifts from human demography (a ~20-year, non-parallelizable reproduction technology capped at a few percent per year) to fabrication throughput and energy capture, permitting growth one to two orders of magnitude higher, with hyperbolic episodes when machine researchers raise their own productivity. (iii) Decoupling: output and human welfare separate completely, and the welfare relevance of arbitrarily large GDP collapses into one state variable: the human ownership share \epsilon_t of the corporate network. A golden-rule decoupling theorem sharpens this. At maximal growth the interest rate equals the growth rate (r = g), so any positive human consumption rate out of wealth makes \epsilon_t decay exponentially at exactly that rate. The human share survives only if the machine economy runs strictly inside its expansion frontier, or if law forces it to. We characterize three terminal regimes – rentier post-scarcity, full circular decoupling, socialized ownership – and the instruments that select among them. The conclusion is narrow: in a post-AGI economy, employment policy is obsolete and ownership policy is everything.
[AI-80] Multi-Method Causal Evidence Synthesis: Ranking Candidate Drivers by Convergent Cross-Method Evidence from Observational Data
链接: https://arxiv.org/abs/2608.20187
作者: Manish Gupta,Dipanjan De
类目: Methodology (stat.ME); Artificial Intelligence (cs.AI)
备注: 36 pages, 4 figures, 17 tables. Reference implementation available from the authors on request
Abstract:Practitioners inferring causality from observational data usually rely on a single method and treat its output as causal truth. Recent tools select an optimal method for a dataset, and recent ensembles aggregate multiple causal-discovery algorithms into one graph, but little work pools evidence across different mathematical traditions, including non-causal ones. We present Multi-Method Causal Evidence Synthesis (MCES), a framework that ranks which candidate drivers in an observational system are most likely relevant to a set of outcomes, and with what strength of evidence. MCES runs eleven methods across eight mathematical traditions on observational panel data and pools their outputs into a Convergent Evidence Score (CES), a linear opinion pool. CES quantifies convergence of evidence across analytical lenses: the degree to which methods with different assumptions point to the same driver-outcome relationship. It does not claim causal identification in the interventionist sense; it supports hypothesis prioritization, not a transferable probability of causation. MCES first applies Structural-Behavioral Decomposition to remove definitional (algebraic) relationships, then runs all methods, normalizes outputs to [0,1], and pools them. We distinguish MCES from method selection, structural ensembles, prediction ensembles, and literature synthesis. Using synthetic data with embedded ground truth, the Sachs protein-signaling benchmark, six Bayesian-network structure benchmarks, and two further synthetic domains, we show MCES ranks true edges near the top (Precision@5 = 1.0, Precision@10 = 0.96 on the primary scenario), with a low empirical rate of null pairs reaching Moderate-or-higher convergence. Our central point is not that the pool beats every individual method, but that no single method is uniformly best across the evaluated scenarios, so MCES offers a method-agnostic default.
[AI-81] Listening Forward: Next Patch Embedding Prediction Enables Scalable Audio Learners
链接: https://arxiv.org/abs/2608.19863
作者: Umberto Cappellazzo,Xubo Liu,Stavros Petridis,Maja Pantic
类目: Audio and Speech Processing (eess.AS); Artificial Intelligence (cs.AI); Sound (cs.SD)
备注: Project website: this https URL
Abstract:Self-supervised learning (SSL) has driven substantial progress in audio representation learning, though existing methods have increasingly relied on elaborate pre-training recipes to reach competitive performance. A markedly different pre-training philosophy underpins the most influential progress in language modeling and, more recently, in visual representation learning: rather than train encoders as static feature extractors, models are trained to predict the next element, a discrete token or a continuous embedding, from the preceding context. Autoregressive prediction thereby provides a unified pre-training interface that transfers across modalities, compelling the model to learn the underlying data distribution. We ask whether such a simple causal paradigm can yield strong audio learners, given that audio’s temporal structure makes autoregressive prediction of patch embeddings a natural fit. We introduce NAPE (Next-Audio-Patch-Embedding prediction), a self-supervised framework in which a causal Transformer predicts each next patch embedding of a log-mel spectrogram from the previous ones, using causal masking and stop-gradient as its sole training signal. The design is intentionally minimalist, avoiding reconstruction decoders, acoustic tokenizers, student-teacher setups, and auxiliary regularization losses. Across six audio and speech benchmarks, NAPE achieves state-of-the-art fine-tuning performance on several tasks, scales consistently across encoder sizes, and yields strong linear-probing results. NAPE also produces structured attention patterns without explicit supervision.
[AI-82] An Irreducible Quantum Advantage in Aligning World Models with Reality
链接: https://arxiv.org/abs/2608.19779
作者: Josep Lumbreras,Hailan Ma,Jayne Thompson,Mile Gu
类目: Quantum Physics (quant-ph); Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
备注: 31 pages, 4 figures
Abstract:World models provide digital simulacra of the true world, allowing agents to be trained and tested before costly real-world deployment. At each time step, they receive an action and generate an observation and reward matching the statistics of the true world. In complex environments where present outcomes depend on events far in the past, this requires memory. One might expect that, by increasing memory, we can always build a model accurately enough to align the optimal agent policies of the real and virtual worlds. We show that this is false for classical world models, even when the true world itself is classical. We construct true worlds for which every finite classical model fails along the same possible trajectory: it either loses the ability to distinguish actions when the true world clearly prefers one, or repeatedly assigns the highest expected reward to suboptimal actions. Its expected-reward estimates also retain a nonvanishing average error. In contrast, each such true world admits a quantum world model using a single qutrit that reproduces it exactly: its reward estimates and preferred actions always match those of the true world, ensuring that the optimal policies of the real and virtual worlds remain perfectly aligned.
[AI-83] Concentrated Liquidity Provision: a Reinforcement Learning Perspective
链接: https://arxiv.org/abs/2608.19389
作者: Georgios Chionas,Charalampos Kleitsikas,Stefanos Leonardos,Leandro Sánchez-Betancourt,Carmine Ventre
类目: Trading and Market Microstructure (q-fin.TR); Artificial Intelligence (cs.AI); Machine Learning (cs.LG); Computational Finance (q-fin.CP); Mathematical Finance (q-fin.MF)
备注: 8 pages, 6 figures
Abstract:Automated market makers (AMMs) are a cornerstone of decentralised finance (DeFi). Constant product markets with concentrated liquidity, such as UniswapV3, are now a well-established design. In these markets, liquidity providers (LPs) face a sequential decision problem: they must decide when to rebalance their positions and which price ranges to allocate capital to as market conditions evolve. We formulate dynamic liquidity provision as a stochastic impulse control problem and use reinforcement learning (RL) to solve it, focusing on providing interpretable solutions. We show that learned policies exhibit rich state-dependent behaviour, allocating liquidity according to mispricing, rebalancing costs, uncertainty, inventory exposure, and heterogeneous risk preferences. These behaviours help compress the left tail of the Profit and Loss (PnL) distribution and avoid catastrophic outcomes under high uncertainty. Finally, we benchmark the RL agents against baseline and sophisticated agents from the AMM microstructure literature and analyse their performance.
[AI-84] Causal Inference under Interference with Learned Exposure Mappings
链接: https://arxiv.org/abs/2608.19224
作者: Cong Cao
类目: Methodology (stat.ME); Artificial Intelligence (cs.AI); Computers and Society (cs.CY); Machine Learning (cs.LG); Applications (stat.AP)
备注: 20 pages, 6 figures
Abstract:Exposure mappings are often assumed to be known in causal spillover analyses. In environmental settings, however, they are typically induced by transport processes that are not directly observed and must instead be learned from pollution data. We study how uncertainty in learned transport processes propagates into exposure mappings and downstream spillover inference under interference. We compare mechanistic transport models with modern operator-learning approaches, including PDE, PINO, FNO, and GeoPT, using both simulation studies and an empirical analysis of California PM _2.5 data. In simulations, all four transport models achieved nearly identical pollution prediction accuracy, yet estimated spillover effects ranged from 1.78 to 2.27. Models that more accurately recovered the induced exposure mapping also produced spillover estimates closer to the true effect. Disagreement was modest for regional interventions but substantially larger for localized point-source interventions. The California analysis showed the same pattern: competing transport models produced similar predictions of observed PM 2.5 concentrations while implying different spillover effects under hypothetical pollution-control interventions. Our findings suggest that predictive agreement alone is insufficient for reliable causal inference when exposure mappings are learned rather than directly observed.
机器学习
[LG-0] A comparison between ceiling-mounted FMCW IR-UWB and Wi-Fi radar for in-bedroom human activity monitoring and sleep interruption detection
链接: https://arxiv.org/abs/2608.20322
作者: Anton Lambrecht,Reda El Hail,Xianjun Jiao,Pieter Crombez,Dominique Schreurs,Peter Karsmakers,Adnan Shahid,Eli De Poorter
类目: Machine Learning (cs.LG)
*备注: This paper has been submitted to IEEE Access Journal and is currently undergoing review
Abstract:Despite their growing importance for contact-free radio frequency (RF) based healthcare monitoring, different radio technologies such as frequency-modulated continuous wave (FMCW) radar, impulse radio ultra-wideband (IR-UWB), and Wi-Fi sensing are rarely compared under identical deployment conditions, as existing studies typically differ in hardware, datasets, and evaluation methodologies. In addition, the performance of ceiling-mounted radars, despite their practical deployment and cost advantages in healthcare environments, remain underexplored. Therefore, this paper presents a controlled comparison and analysis of ceiling-mounted FMCW, IR-UWB, and Wi-Fi sensing using synchronized recordings from 20 participants across six room layouts. All technologies are evaluated with the same convolutional neural network (CNN) on both a fine-grained 10-class human activity recognition (HAR) task and a coarse 4-class sleep monitoring task. IR-UWB achieves the highest cross-subject activity recognition performance (89.0% macro F1), while FMCW generalizes best to unseen room layouts (83.8% macro F1). For sleep monitoring, all technologies exceed 92% macro F1 in unseen environments. The results reveal a fundamental trade-off between recognition performance and environmental robustness, which can be explained through differences in range resolution, antenna diversity, Doppler resolution, and spatial information retention. These findings provide practical guidelines for the design of healthcare-oriented RF sensing systems.
[LG-1] Explainable Transformer Models for Clinical Prediction Tasks on Structured Electronic Health Records
链接: https://arxiv.org/abs/2608.20315
作者: Jun Ni Du,Lukas Adamek,Maxim Kryukov,Flavio Dormont,Ziv Bar-Joseph,Sven Jager,Brandon Rufino
类目: Machine Learning (cs.LG)
*备注: Accepted at MLHC 2026; to appear in Proceedings of Machine Learning Research (PMLR)
Abstract:Predictive models over structured electronic health records (EHRs) remain central to machine learning for healthcare, but few have jointly emphasized quantitative laboratory information and interpretability with respect to input medical events. We present BERT-LER, a BERT-style model for coded EHR timelines pretrained and fine-tuned from a de-identified EHR dataset of 75 million patients, that encodes laboratory test results as discrete tokens while retaining graded information through percentile-based binning, paired with Integrated Gradients for token-level attributions grounded in the input EHR sequence. We benchmark our approach on the public EHRShot benchmark suite and on an asthma severity progression study based on real-world data. This addresses a methodological gap in EHR foundation-style modeling by unifying laboratory value representation and explainability in a single framework, while assessing whether both predictive performance and explanations generalize beyond standard clinical prediction tasks. Across EHRShot and asthma tasks, BERT-LER achieves predictive performance that is competitive with, and on laboratory-related tasks often exceeds, publicly available benchmark models, and provides attributions that align with clinically known risk factors. Our architecture and explainability approach can be applied to many therapeutic areas and prediction tasks using language models trained on structured EHRs.
[LG-2] Physical-Support Confidence Sets for Highly Coherent Dictionaries
链接: https://arxiv.org/abs/2608.20295
作者: Guan-Ju Peng
类目: Machine Learning (cs.LG); Signal Processing (eess.SP); Statistics Theory (math.ST)
*备注:
Abstract:Sparse pursuit after dictionary learning can yield a precise atom support even when its physical interpretation is not justified by the calibration data, especially for highly coherent dictionaries where alternative calibration-compatible dictionaries may assign different physical meanings to the same selected support. We develop resolution-aware physical-support inference that jointly accounts for uncertainty in the learned dictionary and in the representation of a deployment signal. Our cross-dictionary confidence correspondence retains calibration-compatible dictionaries and deployment-compatible sparse representations, then projects the surviving explanations onto physical-support space. For local coherent-atom classes with separation scale s, once the deployment data resolve the coherent-block explanation and its atom support, the minimax physical resolution from N calibration signals satisfies \delta_\mathrmopt(N,s)\asymp\min\s,\frac1\sqrtNs^2\ , with relative resolution governed by the orientation-information scale Ns^6 . Deployment replication improves physical localization only when orientation changes cannot be absorbed by adjusting the active coefficients. For computation, we introduce active endpoint bracketing (AEB), an adaptive finite-bank procedure that evaluates only candidates that can still affect the physical report and otherwise safely coarsens or abstains. Finite-bank experiments, including a four-region synthetic application, show that a point-valued plug-in selector can be physically overprecise, whereas AEB avoids unsupported refinement with fewer candidate evaluations.
[LG-3] Dynamic Structural Causal Modeling for Sleep
链接: https://arxiv.org/abs/2608.20285
作者: Ranveer Singh,Saurabh Mathur,Pranuthi Tenali,Arun Badi,Sriraam Natarajan
类目: Machine Learning (cs.LG)
*备注:
Abstract:The causal dynamics of sleep-disordered breathing are complex and vary across patient populations, hindering the development of targeted interventions. We learn dynamic causal graphs of sleep-disordered breathing from Home Sleep Apnea Test (HSAT) recordings, revealing systematic differences in causal structure across sex and age subcohorts. We do so using the PCMCI+ algorithm on windowed fractional variables derived from 105 HSAT recordings, exploiting domain knowledge via edge blacklisting and employing bootstrap aggregation to address small subcohort sizes. The learned graphs show that temporal self-dependencies and the apnea-desaturation relationship persist across all cohorts, while other relationships vary substantially.
[LG-4] Which Eviction Policy Should an LLM Cache Use? A Systematic Study Across Workloads Capacities and Encoders
链接: https://arxiv.org/abs/2608.20280
作者: Yash Kulkarni,Shubham Harkare,Arvind Suresh Yogesh Babu
类目: Databases (cs.DB); Machine Learning (cs.LG)
*备注: 11 pages, 9 figures
Abstract:Semantic caches reuse an LLM response when the incoming query embedding lies near a cached query, but proposed eviction policies have rarely been compared under one protocol. Using CLEVER, we evaluate FIFO, LRU, LFU, ARC, GDSF, a single-pass streaming adaptation of SISO, and a semantic-redundancy policy across three ordered, deduplicated query corpora, three cache capacities, and two encoders. No evaluated policy improves on LFU by more than 0.041 percentage points in any of the eighteen settings. Replacement is not irrelevant: FIFO and streaming SISO trail LFU by as much as 8.67 and 8.55 points, respectively, at tight capacity. We explain the missing upside with a conditional packing result. Under exact lookup and insert-on-miss, a newly inserted entry cannot have a resident neighbor within the hit radius, so a geometry-aware eviction rule receives little new redundancy signal. A separate audit exposes a larger problem with the evaluated operating point. At MiniLM’s median nearest-neighbor threshold, only 2.1-3.9% of sampled LMSYS and QQP hits are judged answer-substitutable, reducing raw hit rates of 51-60% to quality-adjusted rates of 1.1-2.2%. The cross-encoder study further shows that thresholds do not transfer between embedding models. LFU is the strongest simple default in this protocol; deployment decisions should first establish answer validity and then test sub-point policy differences with exact search. Comments: 11 pages, 9 figures Subjects: Databases (cs.DB); Machine Learning (cs.LG) Cite as: arXiv:2608.20280 [cs.DB] (or arXiv:2608.20280v1 [cs.DB] for this version) https://doi.org/10.48550/arXiv.2608.20280 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
[LG-5] DICS: Data-Informed Centroid Splitting for Decision Tree Classifiers
链接: https://arxiv.org/abs/2608.20258
作者: MD Saifur Rahman Mazumder,Feng Yu
类目: Machine Learning (cs.LG); Machine Learning (stat.ML)
*备注:
Abstract:Decision tree-based models are widely used in machine learning due to their interpretability and strong empirical performance. However, training decision trees can be computationally expensive, particularly for large and high-dimensional datasets, largely due to the exhaustive search over candidate splits at each node. To improve computational efficiency, we propose Data-Informed Centroid Splitting (DICS), a clustering-based framework that constructs a compact and informative set of candidate splits using data-driven priors. By incorporating class-aware structure, DICS significantly reduces the split search space for classification tasks while preserving predictive performance. We further provide theoretical analysis showing that under the stated assumptions, DICS does not degrade the performance of classification trees compared to exhaustive split search. DICS can be incorporated into classification trees, random forests, and gradient-boosting models. Extensive experiments demonstrate that DICS achieves comparable accuracy while substantially reducing training time across synthetic and benchmark datasets, highlighting the benefit of integrating data-informed priors into split selection for scalable classification tree learning.
[LG-6] Decoding silent reading from non-invasive EEG
链接: https://arxiv.org/abs/2608.20186
作者: Ingo Marquardt,Anthilia Alchanat,Priyanka Jain
类目: Machine Learning (cs.LG); Neurons and Cognition (q-bio.NC)
*备注: 45 pages (including 12 pages of Supplementary Material), 11 figures (including 2 in Supplementary Material)
Abstract:Non-invasive decoding of inner speech faces a fundamental data problem: a corpus pairing brain activity with a person’s spontaneous inner monologue cannot be collected, and the available proxy paradigms (cued repetitive and retrospectively reported generative inner speech) are slow to acquire, poorly time-locked, and subject compliance is unverifiable. We therefore treat silent reading as a scalable proxy task and ask how much lexical and semantic information a contrastive decoder can extract from it. We report an open-vocabulary analysis of approximately 240,000 word presentations recorded from a single densely-sampled participant across 393 runs (ca. 49 h) of 19-channel dry-electrode EEG. Words from continuous narrative text were presented in rapid serial visual presentation, with typography randomised on every trial to partially decorrelate word identity from low-level visual form. A convolutional EEG encoder, optionally followed by a causal transformer, was trained with a CLIP-style contrastive objective to align short EEG windows with hidden-state embeddings of the presented word taken from a large language model. Decoding, evaluated as word-grouped top-10 retrieval against permutation baselines, was reliably above chance, extended to mid-frequency and rare words, and scaled log-linearly with training-data volume with no sign of saturation. Removing occipital and posterior-temporal electrodes reduced the word-level gain by roughly one third but left context tracking unchanged. Control analyses separate word-level decoding from narrative context tracking and from a non-neural positional prior introduced by the transformer’s positional embedding. These results establish that open-vocabulary word-level information is recoverable from EEG during silent reading, and that decoding is data-limited rather than saturated.
[LG-7] Exact Algebraic Computation of Learning Coefficients for Two-Dimensional Singular Models
链接: https://arxiv.org/abs/2608.20183
作者: Grégoire Sergeant-Perthuis(1),Elias Tsigaridas(2),Jules Tsukahara(2) ((1) CQSB, Sorbonne Université, (2) Ouragan Team, INRIA)
类目: Machine Learning (cs.LG); Symbolic Computation (cs.SC); Algebraic Geometry (math.AG); Machine Learning (stat.ML)
*备注:
Abstract:Classical information criteria such as the Bayesian Information Criterion (BIC) rely on regularity assumptions that break down for singular models, leading to incorrect model selection in settings such as deep learning. The Widely Applicable Bayesian Information Criterion (WBIC) relies on local learning coefficients \lambda , which in the analytic case coincides with local Real Log Canonical Thresholds (RLCT) of the Kullback-Leibler divergence of the model, to capture correct marginal likelihood asymptotics. Exact computation of the learning coefficients has been limited to special cases, and only sampling-based estimation methods are generally applicable. We present the first deterministic algorithm that computes local RLCTs exactly for any two-dimensional model whose Kullback-Leibler distance is contact equivalent to a polynomial, derive a bound on its complexity, and demonstrate its effectiveness for a broad class of models, with applications including polynomial neural networks. Beyond providing ground truth to calibrate sampling-based estimators, exact computation reveals algebraic structure in learning coefficients that sampling cannot and out-speeds it in the shallow regime.
[LG-8] Ask Self Ask Others: Relation Is All You Need
链接: https://arxiv.org/abs/2608.20172
作者: Yuting Ge,Pengju Yang,Mingkai Nie
类目: Machine Learning (cs.LG)
*备注:
Abstract:Attention directly derives normalized information flow from pairwise scores. We introduce Relation, an alternative token-mixing primitive that first organizes pairwise evidence into explicit Self and Exchange relations and derives information flow afterward. This relational organization gives rise to Full Relation, FlashRelation, Linear Relation, Hybrid Relation, and a KV-style Relation Cache. Across matched decoder-only models at approximately 10M, 30M, and 100M parameters, Full Relation achieves lower final validation NLL than MHA at all three scales. In a fixed-context reference benchmark, FlashRelation is 3.60-4.41x faster than the materialized Full Relation implementation. Across scale-matched production workloads, it reaches 76.4-84.9% of PyTorch FlashAttention throughput while executing the Full Relation operator. Hybrid Relation uses 75% Linear Relation layers and achieves strong language-modeling quality. These results support a relation-first view of token mixing: ask Self, ask Others, then let Flow follow Relation.
[LG-9] Evaluating Neural Cartographic Relief Shading for Urban Environments: A Downtown Calgary Study Using High-Resolution DEM and DSM Data
链接: https://arxiv.org/abs/2608.20149
作者: Emmanuel Stefanakis
类目: Machine Learning (cs.LG)
*备注: 15 pages, 16 figures, submitted for publication
Abstract:This article explores the performance of analytical and neural-based hillshading methods in a dense urban environment using high-resolution digital elevation model (DEM) and digital surface model (DSM) data for downtown Calgary. The study compares single-direction and multi-direction analytical hillshading with relief shading generated in Eduard, a machine-learning system originally developed to emulate Swiss-style shaded relief trained primarily on mountainous landscapes. Because Eduard was not designed for buildings, bridges, streets, trees, and other urban infrastructures, the central question is not whether it perfectly reproduces urban morphology, but whether parameter tuning can nevertheless produce visually strong, cartographically useful, and in some cases superior results when compared with conventional analytical methods. The analysis focuses especially on terrain type, micro and macro generalization, and flat-area detail parameters, while keeping the large-scale shading style constant throughout the neural experiments. The article is structured as an exploratory comparison rather than a benchmark of universal best practice. It aims to identify where analytical hillshading remains more reliable, where Eduard offers unexpected strengths, and where neural shading fails because of its training bias toward alpine terrain. The study contributes to current work on terrain representation by testing whether a neural approach designed for natural landforms can be adapted to a highly built urban setting, and it concludes by arguing for future model training and evaluation specifically targeted at urban relief shading.
[LG-10] SAE-Xplainers: Rule-Based Feature Interpretation for Extreme Earth Events ECCV2026
链接: https://arxiv.org/abs/2608.20117
作者: Hugo Porta,Emanuele Dalsasso,Chang Xu,Theo Gnassounou,Devis Tuia
类目: Machine Learning (cs.LG)
*备注: 22 pages, 16 Figures, Under Review. A non-archival 2-page version was accepted as an oral presentation at Climate Informatics 2026 (Extended Abstract ID 66, this https URL ), and a non-archival 4-page version was accepted as an oral presentation at the AICC Workshop at ECCV 2026
Abstract:The emergence of large-scale Weather and Climate (WC) datasets offers new opportunities for modeling extreme Earth events (ExEE) and their impacts using deep learning. However, their adoption in operational settings remains limited by the lack of models’ interpretability. While for conventional text and image modalities, tools such as Sparse Autoencoders (SAEs) have proven effective for extracting human-understandable concepts, their use for the analysis of ExEE remains challenging due to the nature of WC data. To address this, we introduce (i) a geographic location-based modulation of the inputs of SAE to capture the local semantic meaning of environmental patterns, and (ii) an ensemble of rule-based SAE-Xplainers to interpret the resulting high-dimensional features derived from complex, multi-modal environmental predictors. We evaluate our method on three ExEE types: the prediction of fires, and the detection of tropical cyclones and atmospheric rivers. We show that SAE input modulation improves both reconstruction performance and feature utilization, and that our SAE-Xplainers enable faithful interpretation of complex climatic patterns by unfolding them into human-understandable rules that are consistent with the scientific literature, while also supporting the identification of feature absorption.
[LG-11] Orthogonal JEPA: Factorized Predictive States for Latent World Models
链接: https://arxiv.org/abs/2608.20065
作者: Taoyong Cui,Pheng Ann Heng,Wanli Ouyang
类目: Machine Learning (cs.LG)
*备注:
Abstract:World models construct latent states that support prediction, planning, and reasoning about an underlying system. Joint-embedding predictive architectures (JEPAs) offer a direct way to learn such states by predicting targets in representation space instead of reconstructing every detail of the observation. Standard JEPAs, however, organize all predictable content through one target embedding and one prediction pathway. In complex systems, this monolithic state can allocate redundant capacity to dominant signals while providing weak or conflicting gradients to less dominant predictive structure. We introduce \method, a latent world-modeling framework based on orthogonal predictive factorization. Learned basis matrices analyze each target state into multiple components, and a dedicated prediction branch estimates each component from a shared context representation. Predictive regression preserves the factor magnitudes required for state synthesis, an orthogonality objective discourages repeated directions, factor-activity regularization maintains variation in projected targets, and online variance regularization discourages coordinate-wise encoder collapse. Predicted components are synthesized into a complete latent state that can be used by a readout, decoder, planner, or autoregressive rollout. The same predictive-state mechanism applies when the target is temporally future, spatially hidden, or another partial observation of the same system. Experiments on controlled vision, single-cell transcriptomics, longitudinal health records, continuous control, and molecular dynamics evaluate representation quality, forecasting, planning, and long-horizon stability.
[LG-12] DecoVAE: a Lightweight Interpretable Trend-Seasonal VAE Framework for Efficient Probabilistic Time Series Forecasting
链接: https://arxiv.org/abs/2608.20052
作者: Alexander Marusov,Dmitry Anikin,Alexey Zaytsev
类目: Machine Learning (cs.LG)
*备注:
Abstract:Probabilistic time series forecasting remains challenging, largely because modeling distinct trend and seasonal dynamics requires specialized approaches. Existing methods often fail to capture the unique inner properties of these components, lack interpretability, or suffer from heavy memory and runtime overhead. To address these limitations, we propose DecoVAE, a lightweight interpretable trend-seasonal VAE framework that explicitly decomposes time series into trend and seasonal components by applying domain-specific inductive biases. The trend stream enforces structural smoothness using a differential regularizer on the latent trajectory, analogous to the Hodrick-Prescott filter. Concurrently, the seasonal stream operates in the frequency domain via a complex Gaussian VAE, natively capturing the amplitude and phase of periodic patterns. Extensive evaluations across seven real-world benchmarks show that DecoVAE consistently outperforms strong baselines. It achieves reductions of up to 14.96% in CRPS and 23.30% in NMAE for short-term forecasting, and up to 52.68% and 26.51% for long-term horizons. Crucially, DecoVAE yields these accuracy gains while remaining highly efficient, reducing model weight by up to 93% and accelerating speed by up to 74% compared to the second-best method.
[LG-13] End-to-end Early Classification of Time Series in Non-Stationary Environments
链接: https://arxiv.org/abs/2608.20044
作者: Aurélien Renault,Alexis Bondu,Antoine Cornuéjols,Vincent Lemaire
类目: Machine Learning (cs.LG)
*备注:
Abstract:Early Classification of Time Series (ECTS) requires making accurate decisions as early as possible in inherently online and evolving environments. Yet, most existing methods assume stationarity and rely on separable designs, where classification and triggering are optimized independently, an assumption that fundamentally limits their adaptability under drift. In this work, we challenge this paradigm and study ECTS under non-stationary conditions. We provide the first systematic comparison between separable and end-to-end approaches across controlled drifting scenarios. Building on Reinforcement Learning, we introduce DQeND, a unified architecture that jointly learns representation, classification, and triggering decisions, while remaining directly comparable to state-of-the-art separable baselines. Across a wide range of drifts, DQeND demonstrates strong robustness across various non-stationary scenarios, consistently outperforming separable baselines. An ablation study further highlights that jointly updating representation and decision modules is critical to these gains. Overall, our results indicate that end-to-end learning can offer improved adaptation capabilities for ECTS in dynamic environments, and motivate further investigation of alternatives to separable designs.
[LG-14] CLaST: Context-aware Contrastive VAE for Probabilistic Time Series Forecasting
链接: https://arxiv.org/abs/2608.20025
作者: Alexander Marusov,Dmitry Anikin,Petr Sokerin,Vitaliy Pozdnyakov,Ilya Kuleshov,Alexey Zaytsev
类目: Machine Learning (cs.LG)
*备注:
Abstract:Probabilistic forecasting models are widely used for time series forecasting in domains such as energy systems, finance, medicine, and transportation. In recent years, deep generative models have shown strong results on probabilistic forecasting, yet many conventional approaches struggle to capture internal temporal dependencies, leading to latent representations with limited expressive power. To address this limitation, we propose \textitCLaST, a VAE framework for probabilistic multivariate time series forecasting. Unlike existing generative models, CLaST learns embeddings that preserve contextual similarity between observations through our contrastive loss function. Experiments across nine widely adopted benchmarks demonstrate that CLaST consistently surpasses strong baseline methods. In short-term forecasting tasks, our approach achieves improvements of up to 16.4% in CRPS and 14.4% in NMAE over the second-best method. Furthermore, in long-term prediction CLaST attains superior overall performance, exceeding the second-best method by up to 48.6% and 25.1% in CRPS and NMAE, respectively.
[LG-15] Systematic Evaluation of TabPFN-TS for Zero-Shot Probabilistic Heat Load Forecasting in District Heating Networks
链接: https://arxiv.org/abs/2608.20024
作者: Ben Spoek,Karim K. Ben Hicham,Kai Derzsi,Philipp Althaus,Alexander Mitsos,Dirk Müller
类目: Machine Learning (cs.LG)
*备注: 33 pages, 10 figures; supplementary information included
Abstract:District heating energy hubs require reliable heat load forecasts for efficient operational scheduling. Conventional forecasting workflows train system-specific models on historical data, which can become burdensome when networks change through new consumers, retrofits, or changing operating regimes. Zero-shot time-series foundation models and in-context forecasting offer a promising alternative: they can adapt at inference time from recent observations rather than by repeated retraining. This study systematically evaluates TabPFN-TS against time-series foundation models and trained machine-learning baselines for probabilistic heat load forecasting in district heating networks. Unlike foundation models pretrained on large collections of real time series, TabPFN-TS relies on synthetic pretraining data, which avoids direct pretraining-test overlap but raises the question of whether the learned prior captures district heating dynamics. We analyze covariate choice, context length, temporal resolution, and prediction horizon on representative operating weeks, validate the selected configuration over a full year, and test transferability on a second network. The results identify hourly 24-hour forecasting with a 12-week rolling context and ambient temperature as a parsimonious high-performing configuration; longer context windows do not improve accuracy. TabPFN-TS remains close to Chronos-2 in deterministic accuracy, reaching CVRMSE values of 13.06% versus 12.48% on the main dataset, and lies within the critical-difference threshold in the daily-rank comparison. Although Chronos-2 achieves the lowest aggregate full-year error, TabPFN-TS shows better empirical calibration. Finally, the diagnostic findings motivate a Multi-Resolution Residual-Correction Forecaster that combines a low-frequency Base Forecaster with a short-horizon Residual Forecaster to improve longer-horizon planning accuracy.
[LG-16] Scale-Aware Pretraining of Time Series Foundation Models via Multi-Patch Token Alignment and Hybrid Masking
链接: https://arxiv.org/abs/2608.20005
作者: Taihua Chen,Xiang Ma,Yixin Zhang,Tailin Zhan,Manyu Sun,Lizhen Cui
类目: Machine Learning (cs.LG)
*备注:
Abstract:Pretraining time series foundation models across heterogeneous datasets necessitates effective handling of varying sampling frequencies. Current methods either employ dataset-specific patch sizes and separate FFNs, leading to fragmented representations, or enforce a fixed patch size that neglects inherent temporal variations. To address this, we propose SATS, featuring a scale-aware token alignment mechanism that treats patch size as an explicit notion of scale. By incorporating a contrastive-inspired alignment regularizer, SATS aligns representation spaces across scales while preserving distinct modeling capacities. Furthermore, a hybrid masking strategy combining random and contiguous masking is introduced to capture multi-scale temporal structures. Experimental results on LSTF benchmarks demonstrate that SATS achieves a 9.2% improvement in MSE and an 8.3% gain in GIFT-Eval MASE compared to competitive baselines. Notably, SATS consistently delivers SOTA performance while achieving a 65.6% increase in model efficiency over advanced baselines, highlighting its effectiveness and scalability in time series pretraining.
[LG-17] Green BOA: Determining the environmental break-even point for ML-based data compression
链接: https://arxiv.org/abs/2608.19994
作者: Caterina Doglioni,Akshat Gupta,Thomas Elliott,Hanzila Hussain,Sanjiban Sengupta
类目: Machine Learning (cs.LG); High Energy Physics - Experiment (hep-ex); Computational Physics (physics.comp-ph)
*备注: 3 pages, 1 figure. Accepted as a lightning-talk contribution at the 2nd International Workshop on Low Carbon Computing (LOCO 2026), Lancaster University, United Kingdom, 10-11 September 2026. Part of the LOCO 2026 proceedings, arXiv:LOCO2026/L05
Abstract:We summarise the outcome of two summer internship projects based at the University of Manchester, focused on the break-even point in terms of environmental sustainability for ML-based data compression algorithms. Using the example of a ML-based lossless compression algorithm, we compare estimates for the carbon-equivalent of the infrastructure needed for ML training and inference with the carbon-equivalent savings from reduced disk storage requirements, and discuss their break-even point.
[LG-18] G-MARK: Grounded Multi-Agent Reasoning for Cooperative Driving via Knowledge Graphs ICML
链接: https://arxiv.org/abs/2608.19964
作者: Bhavya Gupta,Onat Gungor,Tajana Rosing
类目: Machine Learning (cs.LG)
*备注: Accepted for oral presentation at the 25th IEEE International Conference on Machine Learning and Applications (ICMLA’26)
Abstract:Autonomous driving systems must operate under partial observability, where safety-critical objects may be occluded or visible only to neighboring connected vehicles. Vehicle-to-vehicle cooperation can reduce this uncertainty, but existing cooperative driving methods often compress multi-agent evidence into latent features or hidden multimodal states. As a result, they obscure which agent observed each object, whether the object is visible to the ego vehicle, and how conflicting evidence affects downstream decisions. We propose G-MARK, a grounded multi-agent reasoning framework that converts cooperative object-centric observations into explicit provenance-aware knowledge graphs (KGs). The resulting KGs preserve object hypotheses together with their source attribution, ego-versus-partner visibility, uncertainty, conflicts, spatial relations, and planning-relevant context. G-MARK then derives a shared feature representation from these KGs, enabling lightweight task heads to support object reasoning, motion prediction, control selection, and trajectory forecasting. Compared with the state-of-the-art baseline, GMARK improves occlusion reasoning accuracy by 42.2%, reduces control-selection error by 13.1%, and achieves comparable trajectory-planning accuracy with a 25.6x smaller structured communication payload. Our code is available at this https URL.
[LG-19] From Noise to Signal: Improving Security Log Anomaly Detection Using LLM s with Endpoint-Specific Logs
链接: https://arxiv.org/abs/2608.19938
作者: Christopher Henshaw,Gour Karmakar
类目: Cryptography and Security (cs.CR); Machine Learning (cs.LG)
*备注: The paper contains 35 pages and 3 figures. The paper has not been submitted or published in any conference or journal. The authors have an aim to publish it in a journal
Abstract:Existing approaches to anomalous behaviour log detection, such as Wazuh rely primarily on predefined detection rules, while statistical anomaly detection approaches such as OpenSearch identify deviations from previously observed behavioural patterns. Recent research has investigated LLMs for log anomaly detection because of their ability to interpret semantic and contextual information. However, LLM-based approaches can be affected by prompt construction, noisy log data, and reliance on generic datasets that may lack endpoint-specific authentication behaviours. To address these limitations, this study develops a standardised instruction-based LLM classification framework for detecting anomalous authentication behaviours, including borderline cases. A controlled cybersecurity testbed was developed to generate endpoint-specific authentication data, producing a curated dataset comprising normal, borderline, and anomalous behavioural scenarios. Three instruction-tuned LLMs, Meta Llama 3.1 8B Instruct, Qwen 2.5 7B Instruct, and GPT-OSS 20B, were evaluated against Wazuh rule-based detection and OpenSearch Anomaly Detection using a common ground-truth severity framework. Meta Llama 3.1 8B Instruct achieved the strongest overall end-to-end detection performance, with an accuracy of 89.3%, recall of 88.2%, F1-score of 91.8%, and false negative rate of 11.8%. In comparison, Wazuh achieved an accuracy of 52.0% and false negative rate of 68.6%, while OpenSearch achieved an accuracy of 49.3% and false negative rate of 74.5%. Meta Llama also detected 80% of the borderline anomalous scenarios, compared with 20% for Wazuh and 15% for OpenSearch. Qwen achieved lower overall detection performance than Meta Llama but recorded the lowest average inference latency and 100% structured-response validity. GPT-OSS demonstrated strong classification performance when valid responses were produced.
[LG-20] Auditing Recorded Predictive Lead Service-Line Classifications Against Physical Verification: A Statewide Study of New York
链接: https://arxiv.org/abs/2608.19922
作者: Muhammad Sarmad Sohail
类目: Machine Learning (cs.LG); Computers and Society (cs.CY)
*备注: 14 pages, 3 figures, 3 tables. Analysis code and data provenance: this https URL
Abstract:Under the US Lead and Copper Rule Revisions, a utility may determine a service line’s material with a predictive model instead of inspecting it. New York State publishes, per address, which method was used. Almost no address carries both a model classification and a physical verification, so the check is between populations within a utility rather than paired addresses. We screen all 153 New York localities that classified at least 100 addresses this way. Seventy-five (49%), covering 125,990 addresses or 57% of those screened, record one value. Zero variance alone is not misconduct: 68 of the 75 match their own verification or have too little to test. Seven are contradicted by their own crews, six beyond any sampling explanation. Five are boroughs of New York City, which file as one system; one is East Rochester, 550 km away. New York City is the largest case: a predictive model is the recorded basis for 43,215 addresses, and on all of them the recorded material is “Known Other”. The city records “Unknown” on 121,779 addresses, 1,880 already excavated, and lead on 120,692. In the model bucket both counts are zero, and the 95% upper bound on the rate is 0.0085%. Across the rest of New York the same method records lead or the hedge “Unknown but could be lead” on 12.21% of 176,888 addresses, a comparison whose weaknesses we report. The model-cleared population is newer, median year built 1984 against 1930, and construction era accounts for about a third of the gap and not the rest: holding era fixed, records-based classification finds lead at 4.3-31.9%, physical verification at 1.5-14.5%, the model in no era. Six era-aware estimators place the expected lead lines among them at 1,150-1,450. Two findings need no comparison: 7,782 of these addresses are in pre-1940 buildings, and the archived 2025 snapshot shows the public-side determination was copied from a customer-side model output.
[LG-21] Multi-Source Wasserstein Distributionally Robust Graph Learning
链接: https://arxiv.org/abs/2608.19914
作者: Chuansen Peng,Yifan Xia,Jinshan Zhong,Xiaojing Shen
类目: Machine Learning (cs.LG)
*备注:
Abstract:Network topology inference from graph signals is central to graph signal processing with applications in neuroscience, sensor, and social networks. In practice, target-domain samples are scarce while heterogeneous source-domain data are abundant. Fusing these sources is challenging: Euclidean averaging works for homogeneous sources but degrades sharply as inter-source divergence grows, collapsing distinct geometries into an inflated, biased consensus. We exploit the Wasserstein metric’s distribution-preserving properties to counter heterogeneity while preserving each source’s intrinsic geometry. We propose MS-WDRO, a multi-source Wasserstein distributionally robust graph learning framework that fuses heterogeneous sources via their weighted Wasserstein barycenter, a geometrically principled nominal distribution, then builds an ambiguity ball around it to hedge residual uncertainty. Minimizing worst-case risk yields a tractable regularized Laplacian estimator solved efficiently via a provably convergent ADMM scheme. We establish non-asymptotic guarantees: a finite-sample concentration bound for the empirical barycenter, a pooling bias lower bound proving naive aggregation is suboptimal, and an out-of-sample excess risk bound decaying at a parametric rate with only logarithmic dependence on source count. To calibrate hyperparameters governing robustness, sparsity, and source fusion, we unroll the solver into a differentiable architecture trained end-to-end, achieving data-adaptive calibration beyond cross-validation while retaining interpretability. Experiments on synthetic benchmarks and the multi-site ABIDE~I neuroimaging dataset show MS-WDRO consistently outperforms seven baselines in graph recovery, sample efficiency, and downstream diagnostic utility, with the largest gains in the sample-scarce regime.
[LG-22] A Layered Simplex Architecture for Large Alphabets
链接: https://arxiv.org/abs/2608.19908
作者: Meir Feder,Yaniv Fogel,Ruediger Urbanke
类目: Information Theory (cs.IT); Machine Learning (cs.LG); Machine Learning (stat.ML)
*备注: 30 pages, 7 figures
Abstract:Probability estimation over large alphabets under log loss is a well-studied problem, with celebrated methods such as the Good-Turing estimator. We introduce and study a new Bayesian estimator with four notable properties. First, its construction is exceptionally simple: multiply independent uniform draws from the probability simplex coordinate-wise and renormalize. Depth is the only structural parameter, and averaging over depths eliminates the need to tune it. Second, the regret of the resulting mixture, the excess code length it pays relative to a code that knows the source, admits an explicit and efficiently computable expression. Third, despite its simplicity and lack of tuned constants, the estimator is competitive across a diverse set of synthetic and real-text benchmarks with substantially more specialized methods, including Good-Turing. Fourth, the tractability of its regret allows us to identify scaling laws in data, alphabet size, and depth. For Zipf targets with exponent above one, the regret has a simple reading as long as the sample reveals only a small fraction of the alphabet. It closely matches the description length of the set of discovered symbols, at one bit of code per bit of description, plus a further cost per symbol. The data exponent is therefore the rate at which new symbols are discovered.
[LG-23] PETA:Parameter-Efficient Test-Time Adaptation for Virtual Screening
链接: https://arxiv.org/abs/2608.19906
作者: Jia-Qi Lin,Yinghua Yao,Chang-Dong Wang,Yew-Soon Ong,Yuangang Pan
类目: Machine Learning (cs.LG)
*备注:
Abstract:Accurately ranking active ligands for a target protein pocket from massive chemical libraries remains a central challenge in virtual screening. DrugCLIP and its recent extensions substantially accelerate this process by encoding protein pockets and molecules into a shared embedding space. Despite this progress, further performance improvements typically require retraining the entire model, incurring substantial computational overhead and making target-specific customization inefficient. In this work, we formulate the specialization of pretrained virtual screening models to individual pockets as a test-time adaptation problem and propose PETA, a parameter-efficient framework that directly adapts pretrained model at test time. Given a target pocket, PETA constructs pocket-specific negatives through molecular diffusion and chemical validity filtering, and further moves them toward the reference ligand retrieved from structural databases via embedding-space mixup to create more challenging ranking tasks. A ranking objective then places greater emphasis on suppressing high-scoring invalid candidates that could contaminate the top-ranked screening results, providing structured supervision for lightweight adaptation. Experiments across diverse benchmarks demonstrate that this lightweight, pocket-specific adaptation outperforms both pretrained and fully retrained baselines while updating only the LayerNorm parameters, which account for approximately 0.03% of the full model.
[LG-24] Reliable Neural Collapse Approximation for Open-World Test-Time Adaptation
链接: https://arxiv.org/abs/2608.19890
作者: Jia-Qi Lin,Yuangang Pan,Chang-Dong Wang,Haizhang Zhang,Ivor W. Tsang,Joey Tianyi Zhou
类目: Machine Learning (cs.LG)
*备注:
Abstract:Test-Time Adaptation (TTA) methods aim to bridge the domain gap between the source and target domains. However, traditional TTA methods become ineffective when the label distribution shift occurs, a challenge commonly referred to as an open-world scenario. In this paper, we introduce a new method named Reliable Neural Collapse approximation (ReNC) for Open-World Test-Time Adaptation (OWTTA). Specifically, we leverage neural collapse as a structural prior for reliable target-domain adaptation. Guided by this prior, we justify that the pre-trained classifier weights can serve as the prototypes of the source domain. By measuring the similarity between samples and prototypes, we filter out the Out-Of-Distribution~(OOD) samples for reliable updates. Furthermore, we propose a neural collapse approximation mechanism to refine these prototypes, ensuring they can gradually adapt to the target domain while maintaining the neural collapse structure. Extensive experiments on several open-world benchmarks demonstrate the superiority of the proposed method. Our empirical analysis suggests that ReNC better preserves NC-related properties in the target domain, providing useful evidence for explaining reliable OWTTA and offering new insights for model design. Code is available at this https URL.
[LG-25] Online Test-Time Adaptation for Generalizable Dynamic Graph Anomaly Detection
链接: https://arxiv.org/abs/2608.19858
作者: Jialun Zheng,Hanchen Yang,Jiannong Cao,Yankai Chen,Yuanjing Feng,Philip S. Yu
类目: Machine Learning (cs.LG)
*备注:
Abstract:Generalizable dynamic graph anomaly detection (DGAD) enables pretrained detectors to identify anomalies in unseen target domains without costly retraining. However, existing methods often fail for two reasons. First, they mainly rely on domain-agnostic patterns and miss domain-specific patterns that keep evolving. Second, they assume access to the full target domain data, whereas in more practical online test-time adaptation settings, target data arrive sequentially in unlabeled chunks. To address these limitations, we formulate online test-time adaptation for generalizable DGAD and propose OTTA-DGAD. OTTA-DGAD first extracts dynamic prototypes, i.e., evolving representations of normal and anomalous patterns, from temporal ego-graphs and stores them in a memory buffer. The buffer selectively retains general patterns shared across the source domains used for pretraining while incorporating new patterns from the target domain. An anomaly scorer then compares incoming edge representations against these prototypes to identify both general and domain-specific anomalies. During adaptation, OTTA-DGAD updates the memory buffer using reliable pseudo-labels identified through confidence-based detection. It further enriches each target chunk with relevant representations retained from previous chunks, compensating for information loss resulting from the sequential arrival of data. Extensive experiments under strict test-then-adapt OTTA settings demonstrate state-of-the-art performance on ten real-world datasets from diverse domains.
[LG-26] Inadvertent Context Leakage in Language Models
链接: https://arxiv.org/abs/2608.19857
作者: Jaiden Fairoze,Neal Mangaokar,Kamalika Chaudhuri,Sanjam Garg,Saeed Mahloujifar
类目: Machine Learning (cs.LG); Cryptography and Security (cs.CR)
*备注:
Abstract:For AI agents to be useful beyond simple chat, they must hold sensitive user context such as calendars, credentials, health records, and financial data. We study whether the mere presence of such secrets in a model’s context window introduces hidden correlations into the model’s benign outputs, allowing reconstruction even when the model correctly refuses direct extraction. We further study whether an adversary can actively engineer prompts that amplify this effect, using the model as a covert carrier to transmit secrets through seemingly innocuous text. In both cases, this limited leakage is exploited using a novel adaptive attack that assumes black-box access to the underlying model. In controlled experiments across eight proprietary models, we find that 2-digit in-context secrets are reconstructed with near-perfect accuracy and 4-digit secrets at 82% exact match, all from outputs the model produces in response to ordinary, non-adversarial requests. We observe that more capable models leak more: stronger instruction-following amplifies sensitivity to in-context secrets, suggesting leakage is a byproduct of capability as opposed to a patchable bug. We show this leakage enables two practical attacks: (1) a trained classifier that infers semantic predicates about user memories (e.g., health conditions, financial events) from routine natural-language outputs, and (2) an RL-trained adversary that extracts full Social Security Numbers from a production-style agent. Subjects: Machine Learning (cs.LG); Cryptography and Security (cs.CR) Cite as: arXiv:2608.19857 [cs.LG] (or arXiv:2608.19857v1 [cs.LG] for this version) https://doi.org/10.48550/arXiv.2608.19857 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
[LG-27] FAR-DPO: Feasibility-Aware and Robust Direct Preference Optimization for Cyclic Peptide Design
链接: https://arxiv.org/abs/2608.19808
作者: Guofeng Zhang,Rong Han,Xiaoyu Wang,Zhiyun Li,Zongbo Han,Xiaohong Liu,Guangyu Wang
类目: Machine Learning (cs.LG)
*备注: 16 pages, 6 figures, and 7 tables. Includes supplementary materials
Abstract:Cyclic peptides are emerging as promising molecular scaffolds in drug discovery due to their high binding affinity and structural stability. However, extending generative models from linear to cyclic peptide design remains challenging, as cyclization sharply restricts the feasible design space through coupled geometric and biophysical constraints. Moreover, limited training data has led existing approaches to rely largely on zero-shot generation or post hoc filtering, resulting in low yields of feasible designs and limited control over multi-objective trade-offs. To address these limitations, we propose FAR-DPO (Feasibility-Aware and Robust Direct Preference Optimization), an architecture-agnostic framework that steers generative models toward structurally and biophysically feasible cyclic peptide designs, particularly for challenging targets. FAR-DPO integrates feasibility-aware preference construction with difficulty-aware group-robust optimization. Specifically, it constructs within-target preference pairs through feasibility-gated multi-objective dominance and adaptively reweights predefined difficulty groups according to their current preference losses. On the CPSea LNR benchmark, under a fixed generation budget, FAR-DPO increases overall success rate from 46.89% to 57.79% on PepGLAD and from 47.96% to 49.57% on PepFlow. These gains also extend to the hardest target quartile and are accompanied by more favorable best-per-target binding scores. Together, these results demonstrate FAR-DPO’s effectiveness in improving feasibility and target-wise robustness.
[LG-28] Answer-Level Trust Selection for Physical Vision-Language Reasoning
链接: https://arxiv.org/abs/2608.19807
作者: Rongyu Yu,Ke Niu,Fengxiang He
类目: Machine Learning (cs.LG)
*备注: Preprint
Abstract:Vision-language models (VLMs) can estimate physical quantities such as duration, speed, and acceleration from visual observations, but existing benchmarks primarily assess overall model performance against annotated ground truth. In deployment, a key question is whether an individual prediction can be trusted when its ground truth is unavailable. Self-consistency alone may fail to capture important failure modes: a VLM may produce stable-but-wrong estimates or rely on textual priors rather than visual evidence. We formulate answer-level selective prediction for quantitative physical reasoning and propose Answer-Level Trust Selection (ATS), a post-hoc, model-agnostic framework for accepting or rejecting individual VLM predictions. ATS requires no fine-tuning, auxiliary verifier, or access to the model’s internal logits. Instead, it aggregates eight interpretable behavioral diagnostic scores derived from repeated queries and controlled interventions into a unified trust score. We evaluate ATS in depth on Qwen2.5-VL-7B and across 20 VLM backbones, examining selective performance, diagnostic behavior, and targeted failure modes. Our results show that intervention-based diagnostics help identify stable-but-wrong and prior-tracking predictions that repeated agreement alone may miss. However, improved failure-case rejection can come at the cost of lower retention of correct predictions. ATS therefore complements model-level capability evaluation with answer-level reliability assessment for quantitative VLM predictions. Code will be released upon publication.
[LG-29] Unsupervised Anomaly Detection Using Flow Matching on Tabular Data
链接: https://arxiv.org/abs/2608.19801
作者: Philip Konz,Tejaswini Medi,Margret Keuper
类目: Machine Learning (cs.LG)
*备注:
Abstract:Financial anomaly detection often relies on large unlabeled transaction logs, where anomalous samples may already be present during training. Such training-set contamination violates the clean-normal data assumption underlying many anomaly detection methods. Although flow matching has demonstrated strong performance in generative modeling, its robustness in unsupervised tabular anomaly detection remains underexplored. In this work, we study flow-matching-based anomaly detection under contaminated training data by comparing Time-Conditioned Contraction Matching (TCCM) with Forest-Flow and evaluating multiple anomaly scoring functions. Our results show that the choice of anomaly score is critical. The original single-step Decision score used by TCCM is sensitive to contamination, whereas trajectory-based Deviation and Reconstruction scores provide more stable anomaly signals. With these scores, Forest-Flow becomes competitive with, and in some cases outperforms, TCCM. These findings highlight the importance of anomaly scoring for flow-matching methods in financial anomaly detection under severe class imbalance.
[LG-30] RecPFN: Prior-Fitted Networks for In-Context-Based Recommendations
链接: https://arxiv.org/abs/2608.19735
作者: En Zhi Tan,Jia Xiang Lim,Bryan Lijie Chew,Tze Minh Ng,Benjamin Yan Han Yap
类目: Machine Learning (cs.LG)
*备注: 12 pages, 4 figures, 8 tables
Abstract:We introduce RecPFN, a prior-fitted network that brings in-context learning to sequential recommendation. RecPFN is pretrained entirely on synthetic clickstream environments sampled from a broad structural causal prior, enabling it to amortize Bayesian-style inference from a small support set. At inference, a lightweight decoder-only transformer conditions on a handful of domain sequences and produces next-item predictions for queries in a single forward pass, without any weight updates. Across eight public benchmarks, RecPFN achives state-of-the-art zero-shot performance while remaining strongly competitive with supervised methods in low-compute and low-data regimes. It is deployment-efficient and robust to domain shift, outperforming strong zero-shot baselines that rely on large real-interaction corpora. RecPFN provides a practical path toward generalizable, data-efficient recommenders and opens avenues for richer priors, longer-context ICL, and multimodal extensions. Code for training and evaluation is publicly available at this https URL.
[LG-31] Learning Deterministic and Stochastic Forced Hamiltonian Systems
链接: https://arxiv.org/abs/2608.19688
作者: Benedikt Brantner,Tomasz Tyranowski
类目: Numerical Analysis (math.NA); Machine Learning (cs.LG); Symplectic Geometry (math.SG)
*备注:
Abstract:We develop a geometric framework for learning deterministic and stochastic forced Hamiltonian systems with neural networks. Motivated by the Lagrange-d’Alembert principle and the theory of variational integrators, we introduce the notion of a Lagrange-d’Alembert map and establish a C^r convergence theorem for first-order one-step methods. Building on these results, we propose Generalized Forced Hamiltonian Neural Networks (GFHNNs), a class of structure-preserving neural networks obtained by concatenating Lagrange-d’Alembert-Euler maps, and prove a universal approximation theorem for this architecture. We further extend the framework to parameter-dependent systems, leading to Parametric Generalized Forced Hamiltonian Neural Networks (PGFHNNs). By interpreting the multiple Stratonovich integrals appearing in the Stratonovich-Taylor expansion as parameters, the same framework can be applied to stochastic forced Hamiltonian systems whenever information about the underlying Wiener process is available. Our numerical experiments demonstrate that the proposed geometric architectures provide significantly improved long-time stability and accuracy compared to non-geometric residual neural networks, while requiring substantially less training data to achieve a comparable level of performance.
[LG-32] CacheRoute: Planned Prefix-Affinity Routing for Large-Scale LLM Serving
链接: https://arxiv.org/abs/2608.19677
作者: Huang Cheng
类目: Distributed, Parallel, and Cluster Computing (cs.DC); Machine Learning (cs.LG)
*备注:
Abstract:Prefix caching avoids prefill only when a repeated request returns to a server that still holds the prefix KV. Cache-blind balancing disperses that reuse; fixed affinity preserves it but can overload a server. CacheRoute resolves this tradeoff with a periodic routing plan. It admits high-rate keys to a stable warm set and places their assignments by expected load. Hot keys may use more than one destination, although every key in our primary semi-synthetic aggregate uses exactly one. On Llama-3.3-70B in fp8 across 60 H100 GPUs, CacheRoute sustains 176+/-11 QPS at a 3.5-s p99 SLO, 2.3x the strongest of five baselines. Served KV-cache hit rate rises from 64.1+/-1.3% under cache-blind balancing to 93.2+/-0.5%. A second semi-synthetic aggregate and controlled 8B and burst experiments separate the effects of affinity and placement. Two 32B workloads provide the counterexamples: when affinity recovers too little KV work, its residual load skew reduces or erases the improvement. We therefore recommend gating any deployment with a shadow replay rather than enabling affinity from workload statistics alone.
[LG-33] SAGE-XGBoost: Spatially Augmented Graph Embeddings–Machine Learning Framework for Natural Hazards Susceptibility Mapping under Data Scarcity
链接: https://arxiv.org/abs/2608.19672
作者: Mohammad H. Vahidnia,Ali Pourkarimi
类目: Machine Learning (cs.LG)
*备注:
Abstract:Natural hazard susceptibility mapping is often constrained by limited labeled data, reducing the generalizability of conventional machine learning and limiting the applicability of complex deep learning models. This study proposes SAGE (Spatially Augmented Graph Embeddings), a structurally informed feature-engineering framework that combines controlled noise-based data augmentation with neighborhood-based graph embeddings to improve prediction under data-scarce conditions. A K-nearest neighbor graph is constructed to derive local spatial statistics, which are reduced using principal component analysis and integrated with environmental covariates and spatial coordinates. The resulting features are used with XGBoost to develop the SAGE-XGBoost model. The framework was evaluated for landslide and wildfire susceptibility mapping. SAGE-XGBoost consistently outperformed conventional and spatially explicit machine learning models. Compared with Spatial XGBoost, it achieved an absolute improvement of above 33 percentage points across the two case studies. The model reached AUC values of approximately 0.97 for landslide susceptibility and 0.95 for wildfire susceptibility. Feature importance analysis confirmed the contribution of graph embeddings to prediction, while their integration improved spatial coherence and reduced local noise amplification. Overall, SAGE-XGBoost provides an efficient and transferable alternative to deep representation learning for environmental hazard assessment and other geospatial prediction tasks under limited supervision.
[LG-34] FleetSieve: Decision-Critical Profiling for SLO-Aware LLM Fleet Configuration
链接: https://arxiv.org/abs/2608.19659
作者: Huang Cheng,Scott Zhang,Aubert Li
类目: Machine Learning (cs.LG); Distributed, Parallel, and Cluster Computing (cs.DC)
*备注:
Abstract:Choosing tensor-parallel (TP) degrees and replica counts for an LLM serving fleet is difficult because performance is not monotonic in TP and the feasible choice can change with load. Exhaustive profiling resolves this uncertainty, but measures many configurations that do not affect the final resource allocation. We present FleetSieve, which selects measurements according to their expected effect on a resource-coupled, SLO-aware fleet decision. FleetSieve models capacity and tail latency jointly, compares conservative and optimistic allocations, and stops when their remaining decision gap is below a specified tolerance. On a fixed H100 measurement grid for a 31B-parameter open-weight model, FleetSieve reaches the oracle aggregate decision using 22,200 GPU-seconds, 6.9% less than uniform random profiling in the fixed comparison. Across 200 random reveal orders, its mean saving over random profiling is 5.4% (95% bootstrap CI: 3.5-7.2%). The fixed-comparison saving is 21.5% for Chat, while FleetSieve does not use the fewest GPU-seconds for Code. Joint capacity and tail modeling also avoids selecting a configuration whose 46.4-second completion p99 violates a 30-second SLO. In a 16-GPU allocation, an incorrect sparse-profile decision loses up to 1.93 requests/s and 12.4 percentage points of max-min fulfillment. Boundary repeats and BurstGPT measurements support the observed load-dependent tail-latency mechanism.
[LG-35] Rationally Enriched Chebyshev Trunk Bases for DeepONet Surrogates of High Péclet Entrance Transport
链接: https://arxiv.org/abs/2608.19658
作者: Mingeun Choi,Satish Kumar
类目: Machine Learning (cs.LG); Numerical Analysis (math.NA)
*备注: 39 pages, 9 figures
Abstract:This study demonstrates a rationally enriched Chebyshev (REC) trunk for deep operator network (DeepONet) surrogate models of singularly perturbed and high-Péclet transport problems whose solution profiles are characterized by thin localized boundary or wall layers. The REC trunk combines Chebyshev polynomial dictionary elements with rational dictionary elements constructed using the adaptive Antoulas-Anderson (AAA) algorithm. Over five independent training runs, the resulting REC-trunk DeepONet is evaluated against a vanilla DeepONet and a Chebyshev-trunk DeepONet whose prescribed dictionary consists only of Chebyshev polynomials across three problems whose singular perturbation parameters are diffusion-to-advection ratios: a singularly perturbed scalar boundary-value problem (BVP), the thermal entrance problem with a prescribed wall temperature, and the concentration entrance problem with an absorbing wall. Across the held-out test profiles, the REC-trunk DeepONet improves over the vanilla DeepONet and remains comparable to the Chebyshev-trunk DeepONet in predicting the scalar profile, with its clearest advantage over the Chebyshev-trunk DeepONet appearing when the perturbation parameter lies between 1.00\times10^-4 and 1.78\times10^-4 , where it reduces the profile-error metrics by up to 19.5,% relative to the Chebyshev-trunk DeepONet. In predicting the wall-normal temperature and concentration profiles, the REC-trunk DeepONet reduces the profile-error metrics by up to 60.2,% and 32.2,% relative to the vanilla and Chebyshev-trunk DeepONets, respectively, while suppressing artificial near-wall oscillations as the Péclet or mass-transfer Péclet number ranges from 10^2 to 10^4 .
[LG-36] me-Uniform Self-Normalized Concentration for Discounted Least Squares: Limits and Corrections
链接: https://arxiv.org/abs/2608.19643
作者: Yi-Shan Wu
类目: Machine Learning (cs.LG); Machine Learning (stat.ML)
*备注:
Abstract:Self-normalized concentration inequalities are standard tools in bandit and reinforcement-learning analyses. A widely used weighted extension claims an analogous time-uniform guarantee for discounted least-squares estimators in non-stationary problems. A simple scalar Gaussian counterexample with a fixed parameter shows that the claimed bounded radius is crossed with probability one. For fixed discount and regularization parameters, we further show that, when \delta\leq1/2 and T/\delta is sufficiently large, any deterministic anytime boundary valid uniformly over the stated conditionally sub-Gaussian model class must be at least of order R\sqrt\log(T/\delta) at some time by horizon T ; for nondecreasing boundaries, this order is required at time T . We identify the proof error: different terminal times use different Gaussian mixing distributions, so the fixed-time mixtures do not form one supermartingale, and the stopping-time argument does not repair this failure. Finally, we show that the weighted inequality remains valid at each fixed deterministic time, give valid finite- and infinite-horizon corrections, and discuss consequences for downstream analyses.
[LG-37] Complementary Not Cumulative: Interaction Effects in Physics-Informed Neural Networks for Navier-Stokes Vortex Shedding
链接: https://arxiv.org/abs/2608.19632
作者: Devesh Shah
类目: Machine Learning (cs.LG); Fluid Dynamics (physics.flu-dyn)
*备注:
Abstract:Physics-informed neural networks (PINNs) embed governing partial differential equations directly into the training loss, offering a promising alternative to costly CFD solvers for unsteady flows. Yet the growing list of techniques proposed to improve PINN training is typically validated one at a time, leaving open whether these techniques actually compose. We study this question in depth on the DFG/Schafer-Turek unsteady cylinder wake benchmark. In isolation, nearly every technique performs no better than an untreated baseline. However, combining periodic (SIREN) activations with causal weighting unlocks a previously inaccessible regime, reconstructing velocity and pressure fields to within 4.1% average relative L2 error against an OpenFOAM reference solution. Adding further techniques instead causes catastrophic performance degradation, demonstrating that individually effective PINN interventions can interact nonlinearly and that more elaborate training recipes are not necessarily better.
[LG-38] Unregularized Convergence of Single-Loop Entropy-Regularized Natural Actor-Critic
链接: https://arxiv.org/abs/2608.19587
作者: Zhiqiang Tan
类目: Machine Learning (cs.LG)
*备注:
Abstract:While entropy regularization is widely used to stabilize and accelerate Natural Policy Gradient methods, its ability to yield faster convergence rates for the unregularized objective remains underexplored. Existing analyses often rely on double-loop architectures and invoke a linear entropy penalty. To bridge the gap between theory and practice, we analyze a single-loop, entropy-regularized Natural Actor-Critic algorithm under compatible linear function approximation. By training an uncentered critic, our critic tracking can remain stable even as the training policy approaches determinism and the Fisher information matrix degenerates. We focus on two primary regimes for the optimization landscape: a Stochastic Regime, where we fuse coupled actor-critic updates into a joint Lyapunov recurrence, and a Deterministic Regime, where we pivot to a Policy Mirror Descent framework to circumvent the collapse of Euclidean geometry. By exploiting a positive Minimal Action Gap in the unregularized Markov decision process, we introduce an Exponential Translation mechanism that maps the regularized gap to the unregularized one up to an exponentially decaying tail. By tuning the fixed temperature, our algorithm achieves accelerated unregularized convergence rates, up to approximation-error terms: \tilde\mathcalO(T_total^-1) in the Stochastic Regime, and \tilde\mathcalO(T_total^-2/3) for the average iterate alongside \tilde\mathcalO(T_total^-1/3) for the last iterate in the Deterministic Regime. Here, T_total denotes the total number of stochastic critic updates (or Monte Carlo rollouts). Furthermore, in the tabular setting, our positive-action-gap analysis yields a \tilde\mathcalO(T_total^-2/3) average-iterate rate, surpassing the \mathcalO(T_total^-1/2) worst-case statistical barrier that applies without a positive action margin.
[LG-39] Kähler landscapes for complex neural network descents and guarantees including a search and destroy of the Calabi-Yau manifold
链接: https://arxiv.org/abs/2608.19584
作者: Andrew Gracyk
类目: Machine Learning (cs.LG); Differential Geometry (math.DG); Machine Learning (stat.ML)
*备注: First version
Abstract:We study landscapes for complex-parameterized networks. Our approach is motivated with an information-theoretic manifold perspective of the parameter and via classical optimization guarantees although of complex geometric variety such as through Dolbeault asymptotics. The descent path admits a Kähler information metric under a cross-entropy via the Wirtinger Hessian on the log-likelihood potential. We restrict attention to a descent update rule with natural gradient descent via a differentiated loss scaled by the inverse metric, so the descent path remains in the holomorphic tangent bundle. We emphasize Calabi-Yau information manifolds which profane theoretical guarantees via an ill-curvature-conditioned landscape. Under a Calabi-Yau metric, specifically in a non-compact setting with a global potential so defined geometrically rather than invoking the topological requirements of the Calabi conjecture, a wedged nowhere-vanishing holomorphic form is the top exterior product of the Kähler form up to constants, yielding a constant determinant condition. Under a fixed determinant, a metric almost low rank up to an eigenvalue tolerance implies a blow-up effect. Moreover, it has been discovered that negative curvature subverts the loss landscape, specifically sectional curvature, so we expand on this and draw interconnections to negative-definite Ricci curvature. Our arguments primarily exist in a geometric analytic modality, although we establish roots in deep learning theory such as through asymptotics at initialization and connections through failure modes of neural network guarantees under vanishing and negative Ricci curvature.
[LG-40] A Two-Stage Time-Aware Transformer for Short-Horizon AECOPD Risk Prediction
链接: https://arxiv.org/abs/2608.19578
作者: Dongyang Wang,Weihao Qu,Ling Zheng,Haowen Pan
类目: Machine Learning (cs.LG)
*备注: 7 pages. Accepted for publication in IEEE Systems, Man, and Cybernetics Letters
Abstract:Acute exacerbation of chronic obstructive pulmonary disease (AECOPD) can worsen rapidly, making timely prediction a clinical priority. Most existing machine learning approaches rely on episodically collected clinical variables, introducing delays that limit their practical utility in home monitoring settings. Home ventilators offer a lower-latency alternative, producing a near-continuous record of respiratory status during daily use. However existing ventilator-based approaches either compress the waveform into handcrafted features or focus primarily on binary risk classification, leaving the timing of an impending event unresolved. In this paper, we present a two-stage framework that operates directly on raw pressure and flow waveforms from the most recent seven days of home ventilator use. The first-stage classification model identifies patients at high risk of a severe exacerbation. The second-stage regression model then estimates how many days remain before the event occurs. Our experimental results demonstrate that the two-stage model outperforms traditional baseline models on both risk classification and time-to-event estimation, with our selected Stage 1 classifier achieving F1 = 0.91 and our Stage 2 regression model achieving RMSE = 1.00 days and R^2 = 0.76, giving clinicians both an early warning and actionable lead time before a severe exacerbation occurs.
[LG-41] Empirical Characterization of Learning Geometry in Hybrid Quantum Forecasting Models
链接: https://arxiv.org/abs/2608.19497
作者: Sandra Leticia Juárez-Osorio,Jorge I. Hernandez-Martinez,Jesus Ivan Ruiz-Martinez,Andres Mendez-Vazquez,Eduardo Rodriguez-Tello
类目: Machine Learning (cs.LG)
*备注:
Abstract:We characterize the learning dynamics of a compact hybrid quantum forecasting model through comparison with a structurally aligned classical baseline. Using stationary harmonic-mixture and nonstationary chirp benchmarks with controlled spectral complexity and data availability, we analyze empirical Neural Tangent Kernel dynamics through kernel-target alignment, kernel drift, spectral concentration, and training loss. The classical model exhibits stronger early target alignment, whereas the hybrid model generally develops a less concentrated kernel spectrum and smaller kernel drift. Despite these distinct optimization geometries, both architectures attain similar held-out performance across the evaluated regimes. Notably, the hybrid model uses 125 trainable parameters compared with 281 for the classical baseline and reaches its validation-selected checkpoint earlier in 15 of 18 frequency conditions. A Fourier-augmented classical baseline does not reproduce the observed training behavior, while a controlled re-uploading ablation shows that repeated encoding systematically modifies both optimization and kernel geometry. These results demonstrate that comparable generalization can emerge from substantially different learning trajectories and that individual NTK diagnostics do not provide monotonic predictors of validation convergence. Rather than claiming a general quantum advantage, the study identifies architecture-dependent learning behavior that is masked by endpoint accuracy alone.
[LG-42] Beyond Multimodal Alignment: Certifying Physical Language through Response Substitution and Ordered Execution
链接: https://arxiv.org/abs/2608.19492
作者: Kaizhen Tan,Xin Xu,Siru Tao,Yixiao Li,Hanzhe Hong,Yang Feng,Heqing Du
类目: Machine Learning (cs.LG); Robotics (cs.RO)
*备注:
Abstract:World models increasingly treat compact multimodal representations as interfaces between perception and physical interaction, yet existing probes do not establish whether different sensors carry the same executable meaning or whether that meaning survives a new action composition. We introduce an operational capability hierarchy and the Disjoint-Bridge Operator-Substitution Certificate (DBOSC), which asks whether independently trained modality compilers enter a frozen response chart interchangeably on evidence outside their training panels. On Cluster Haptic, audio and acceleration representations of the same unseen surface are 4.5x closer in response space than wrong-surface pairings, with the gap holding for all 19 held-out surfaces; unsealing withheld responses confirms that every branch predicts the physics better than the population chart. We then test ordered execution in a controlled elastoplastic system with complementary modality blind spots. At the pre-registered budget, the prerequisite refuses the stack because the frozen executor cannot advance even an exact chart coordinate through a held-out program. At a converged budget, the same rank-three chart executes those programs (oracle NMSE 0.18), fusion improves on both modalities, and 14 of 16 registered checks pass; the two failures arise because a diagonal restriction of the fused information matrix performs as well as the full one. Clearing the gate is a property of the executor, not the chart: an executor emitting whole programs instead of shared per-step dynamics is 38x worse than an entity-blind predictor on the same chart. A matching non-identifiability result explains why compression and fusion alone cannot determine an unseen composition law. These results separate attribute access, response substitution, fusion closure, and ordered execution into distinct, separately testable achievements.
[LG-43] When to Retrain: An Empirical Study of Retraining Policies for Streaming ML Under Concept Drift Budget and Latency Constraints
链接: https://arxiv.org/abs/2608.19488
作者: Sawan Dasari
类目: Machine Learning (cs.LG)
*备注: 16 pages, 7 figures, 7 tables. Code and data: this https URL
Abstract:Production machine learning systems degrade under concept drift, yet practitioners have little principled guidance on when to retrain. Retraining is costly, retraining budgets are finite, and a retrained model does not take effect instantly: training and deployment latency leave a stale model serving predictions while the data continues to move. We present a controlled empirical study of three practical model-refresh policies (periodic retraining, error-threshold triggering, and statistical drift-triggered retraining with ADWIN) against a no-retrain baseline, evaluated under a unified system model that makes retraining budgets and training-plus-deployment latency explicit. Across 3,933 experiment runs spanning three drift regimes, three budget levels, up to five latency levels, three datasets, and two learning modes, we find that the single most consequential design decision is not the retraining policy but whether the deployed model learns incrementally. With per-sample incremental updates, and for the linear online learner with immediate labels studied here, no policy differs from the no-retrain baseline by a practically significant margin in any of 54 paired comparisons, even at extreme latency. Without incremental updates, policy choice separates outcomes by 15-55 percentage points of post-drift accuracy, and simple periodic retraining significantly outperforms both reactive policies under abrupt and gradual drift, while reactive policies retain an advantage only under recurring drift. We document systematic failure modes of reactive policies and a latency-budget queueing interaction that silently halves effective retraining budgets, and release the full simulator, dataset pipelines, and per-run artifacts for reproducibility.
[LG-44] LLM as Detector: An In-context Learning Approach for Tabular Anomaly Detection ICONIP
链接: https://arxiv.org/abs/2608.19463
作者: Tu Anh Hoang Nguyen,Dang Nguyen,Thuc Duy Le,Trung Le,Sunil Gupta
类目: Machine Learning (cs.LG)
*备注: Accepted at International Conference on Neural Information Processing (ICONIP) 2026
Abstract:Anomaly detection in tabular data is challenging because abnormal samples often arise as violations of cross-feature dependencies rather than simple marginal deviations. Existing detectors rely on geometric or reconstruction signals, while prior LLM-based approaches mainly fine-tune LLMs with normal samples or generate synthetic anomalies. We propose LLM-Detector, a framework that utilizes the in-context learning capacity of LLMs for structured, prompt-conditioned scoring synthesis, enabling LLMs to derive anomaly detection logic from structured normal-state knowledge. Specifically, normal training data are converted into statistical summaries, causal dependencies, and distilled prototypes that are organized into a prompt for code generation. The resulting scoring engine evaluates statistical deviation, structural inconsistency, and density-based abnormality then computes an anomaly score for each test sample. We evaluate LLM-Detector on 24 tabular datasets, comparing against 15 SOTA baselines. Results show consistent improvements across both mixed-type and continuous-only settings. Moreover, this design eliminates the need for LLM fine-tuning or neural network training, reducing computational cost and enabling practical anomaly detection in real-world tabular systems.
[LG-45] Quantifying Event Impacts on Time Series via Multiscale Contrastive Learning
链接: https://arxiv.org/abs/2608.19447
作者: Yiming Sun,Shengyu Chen,Zhengzhang Chen,Haoyu Wang,Xiaowei Jia,Haifeng Chen
类目: Machine Learning (cs.LG)
*备注:
Abstract:Shocks that spread through the web, such as cybersecurity breach disclosures, can abruptly disrupt financial time series and cause substantial abnormal losses. While these events are disclosed as discrete records through news reports, regulatory filings, or public databases, their consequences unfold through continuous market dynamics. This creates an event-conditioned impact prediction problem: given pre-event market history and limited event metadata, the goal is to estimate short-term post-disclosure abnormal loss rather than reconstruct the full post-event trajectory. However, most time-series forecasting models focus on endogenous regularities such as trend, seasonality, and autocorrelation, and thus struggle with rare and heterogeneous external events. The challenge is further amplified by sparse high-impact events and background market noise. We introduce EventTime, a multi-resolution framework that combines long-horizon market context, short-horizon pre-event dynamics, and event metadata. It incorporates an event fusion module that couples temporal representations with event attributes to identify relevant recent market patterns. To mitigate sparse supervision, EventTime further introduces a dynamic contrastive objective that constructs event- and time-series-aware positive and negative pairs during training. We also construct SECURE, a real-world dataset aligning cybersecurity incidents with stock-market time series and structured and LLM-derived semantic features. Experiments show that EventTime consistently outperforms state-of-the-art time-series and event-aware baselines in estimating post-event financial losses. Further analyses demonstrate more event-sensitive representations, greater robustness to incomplete metadata, and more interpretable estimates of short-term market impact following cybersecurity disclosures.
[LG-46] Data-Driven Time-Varying Control Barrier Functions for Adaptive Safe-Set Learning with Online Decremental Support Vector Machines
链接: https://arxiv.org/abs/2608.19366
作者: Shawon Dey,Michael Budihartono,Hever Moncayo
类目: Computational Engineering, Finance, and Science (cs.CE); Machine Learning (cs.LG)
*备注: 14 pages, 9 figures
Abstract:Mission-critical intelligent systems often operate under time-varying limitations that reduce control authority and change the admissible safe operating envelope. In such settings, a safety certificate learned under nominal conditions may become invalid as system capability changes. To address this challenge, this paper proposes a degradation-aware, data-driven safety-filtering framework that learns a safe set from data, updates it online, and enforces the resulting learned barrier through a time-varying control barrier function (CBF). A nominal safe envelope is first learned from operational data using a radial basis function (RBF)-kernel support vector machine (SVM), whose decision function serves as the initial CBF candidate. To capture capability-induced safe-set contraction, a continuous-time decremental SVM update law is developed so that selected support-vector coefficients are reduced according to a degradation signal. A homotopy-smoothed SVM-CBF is then introduced to avoid discontinuous changes in the learned barrier during active-set transitions. The resulting time-varying learned barrier is enforced using a quadratic-program-based safety filter under degraded input constraints. Forward invariance of the learned time-varying safe set and recursive feasibility of the safety filter are established. Simulation results on a vertical takeoff and landing (VTOL) model show that the proposed method maintains safety under reduced control authority and avoids abrupt barrier-switching effects during safe-set contraction.
[LG-47] Uncovering the Limits of Proof Sharing for Neural Networks
链接: https://arxiv.org/abs/2608.19351
作者: Kanak Das,Shubham Ugare,Bor-Yuh Evan Chang,Sasa Misailovic,Gagandeep Singh,Manu Sridharan
类目: Machine Learning (cs.LG)
*备注: To appear at the 33rd Static Analysis Symposium (SAS 2026)
Abstract:Robustness verification of neural networks is increasingly important, due to their use in many critical domains. In certain scenarios, proof sharing has been shown to accelerate incomplete verification techniques by reusing intermediate-layer abstract states, or templates, across queries. However, questions remain as to the robustness of template-based acceleration across varying network architectures, properties, datasets, and training methods. In this work, we perform a systematic study of the effectiveness of template-based acceleration and its limits. Our study shows that template subsumption rates can vary widely across scenarios. We present a novel metric of jointly stable neurons to explain this variation, showing that in some cases template-based techniques are very unlikely to provide any speedup. Then, we present FastCert, a novel technique for automatically distributing templates across neural network layers to increase performance impact, eschewing templates entirely if they are unlikely to produce a speedup. Across a large set of covering-design based L_0 -verification tasks, FastCert achieved an average speedup of 1.13x over an extant template-based reuse technique.
[LG-48] Holtercare-Bench: A Multimodal Benchmark for Evaluating Long-Term Dynamic ECG Analysis
链接: https://arxiv.org/abs/2608.19297
作者: Yihan Xie,Hanwen Cui,Runze Ye,Juekai Lin,Haoyang Wang,Jinhao Mao,Bo Zhang,Wenqiao Zhang,Xiaogang Guo,Jun Xiao,Lei Zhang
类目: Machine Learning (cs.LG)
*备注:
Abstract:While multimodal large language models (MLLMs) excel in medical applications, most of them favor static images or short-term signals. In the critical field of dynamic electrocardiograms (ECG), models struggle with complex temporal reasoning and diagnostic report generation due to a lack of high-quality datasets and benchmarks. To address this, we introduce (i) Holtercare-23K, a large-scale multimodal dynamic ECG dataset comprising 22,980 QA pairs derived from 788 clinical Holter records and featuring a novel signal-video-text tri-modal alignment. Based on this dataset, we present (ii) Holtercare-Bench, a multimodal benchmark that evaluates models on temporal localization, clinical diagnosis, and global summarization. Zero-shot evaluations of leading MLLMs reveal a significant performance gap in processing ultra-long pathological sequences. However, fine-tuning representative models yields substantial improvements. This work illuminates the limitations of current MLLMs in electrophysiology and provides a foundational benchmark for long-term medical MLLMs. Our project is available at this https URL.
[LG-49] riangular Fuzzy Rescaling Distance
链接: https://arxiv.org/abs/2608.19234
作者: Eddy Soria,Aida Valls,Ana Beatriz Hernández-Lara
类目: Machine Learning (cs.LG)
*备注: 12 pages, 2 tables
Abstract:Decision-making in complex systems often involves dealing with imprecise or uncertain information, frequently represented using fuzzy sets, particularly Triangular Fuzzy Numbers (TFNs). A crucial aspect of many fuzzy methods is the quantification of distance between TFNs. Many distance measures assume that all values are in the same scale, requiring a preliminary normalization stage when applied to heterogeneous attributes with different scales or units. This paper proposes the Triangular Fuzzy Rescaling Distance (d_TR), a metric designed to address this challenge. The d_TR uniquely integrates Linear Rescaling (LRE) directly into the distance calculation, ensuring normalization during the comparison of fuzzy numbers. We formally prove that d_TR satisfies the properties of a metric, including non-negativity, identity, symmetry, and the triangle inequality. Furthermore, we demonstrate that d_TR is bounded, scale-invariant, and origin-invariant. These properties, combined with a weighting vector for prioritizing dimensions, make d_TR suitable for applications involving heterogeneous fuzzy data, such as the construction of synthetic indicators, distance-based machine learning algorithms or multicriteria-decision aiding.
[LG-50] VQC-ZTI: Variational Quantum Control for Zero Trust Protection of the Tactile Internet
链接: https://arxiv.org/abs/2608.18572
作者: Mubassir Serneabat Sudipto(Iowa State University),Shakil Ahmed(Grand Valley State University),Ashfaq Khokhar(Kansas State University)
类目: Cryptography and Security (cs.CR); Machine Learning (cs.LG); Networking and Internet Architecture (cs.NI)
*备注: 7 pages, 3 figures, 3 tables. Accepted at IEEE Global Communications Conference (GLOBECOM 2026)
Abstract:Tactile Internet services couple cyber events directly to physical actuation, so security decisions must improve risk discrimination without perturbing the control path. This paper presents VQC-ZTI, a split-plane Variational Quantum Classifier framework for zero-trust protection of Tactile Internet services, in which an off-path VQC analyzes encrypted-flow telemetry while an on-path policy engine applies cached deterministic grant, restrict, step-up, and deny actions. By decoupling anomaly scoring from enforcement, VQC-ZTI preserves predictable control behavior and allows detector sensitivity and policy aggressiveness to be tuned independently. We evaluate the framework on CESNET-derived aggregated traffic using random, entity-group, and temporal holdouts with a hybrid PyTorch-PennyLane implementation. The full-hybrid Quantum Neural Network achieves mean areas under the receiver operating characteristic curve of 0.9981, 0.9974, and 0.9941 and reduces the false-positive rate relative to ExtraTrees by 44.6%, 49.6%, and 67.9%, respectively. A representative component-timing decomposition further illustrates that batched VQC scoring remains in the asynchronous evidence path rather than the immediate enforcement path.
[LG-51] Information on trajectories: martingales and random times
链接: https://arxiv.org/abs/2608.20337
作者: Akshay Balsubramani
类目: Probability (math.PR); Information Theory (cs.IT); Machine Learning (cs.LG); Statistics Theory (math.ST)
*备注:
Abstract:Accounting for information flow on the path space of trajectories of a nonnegative martingale yields exact variational identities for it, even at arbitrary random times. This recovers the widely used classical concentration inequalities, from Ville to PAC-Bayes, and measures what each one discards. The tail a bound controls is itself a relative entropy, resolved by the chain rule into per-step conditional divergences. The discarded slack has an exact form in each of three geometries: a Gibbs tilt for the Azuma-Hoeffding and PAC-Bayes bounds, the crossing itself for Ville’s and for pooled tests, and a dominating certificate for the L^p maximal bound. That certificate’s optional-stopping deficit resolves per step into Bregman divergences of the running maximum. On a path-time space, the same identity gains one factor that prices anticipation: an arbitrary random time carries an e-process ``peeking penalty.‘’ The partition function can be read as a coalescent–a prefix-sharing probability of independent copies–and geometric mixtures of test martingales gain a pooling benefit for multi-model safe testing.
[LG-52] TCP_α: Margin-Controlled Confidence estimation for reliable Music Information Retrieval
链接: https://arxiv.org/abs/2608.20326
作者: Parampreet Singh,Anushka Singh,Sumit Kumar,Vipul Arora
类目: Audio and Speech Processing (eess.AS); Machine Learning (cs.LG)
*备注:
Abstract:Deep neural networks are often overconfident, assigning high confidence even to incorrect predictions. Consequently, users lack a reliable signal for deciding when a prediction can be trusted. Post-hoc confidence estimation addresses this by training a lightweight auxiliary head over a frozen classifier. Existing targets, however, suffer from inherent ambiguity: they assign overlapping confidence values to correct and incorrect predictions, while errors near the decision boundary receive confidence scores indistinguishable from correct predictions. In this work, we propose TCP_\alpha , a novel confidence target that resolves these limitations by introducing a margin-controlled penalty for misclassified samples. We prove that TCP_\alpha guarantees complete separation between the target values of correct and incorrect predictions, with a separation margin that is independent of the number of classes and increases monotonically with the penalty parameter. Since accurate classifiers naturally produce very few errors, learning these targets results in a severely imbalanced regression problem. We therefore present a systematic study of training strategies for learning under this imbalance and identify an effective training configuration through extensive ablation studies. We evaluate the proposed approach on rāga identification, investigate its robustness under domain shift, and further validate it on frame-wise ornamentation detection without modifying the selected configuration. Across all settings, TCP_\alpha consistently outperforms existing confidence targets for failure prediction. Rejecting only the least-confident 8% of predictions improves the base model’s macro-F1 from 0.89 to 0.98, while fine-tuning the confidence head with only 5% labeled samples from a new corpus effectively restores performance under domain shift.
[LG-53] ransfer Learning in Nonparametric Regression with Deep ReLU Networks ICML2026
链接: https://arxiv.org/abs/2608.20255
作者: Junpeng Ren,Carlos Misael Madrid Padilla,Yanzhen Chen,Oscar Hernan Madrid Padilla
类目: Machine Learning (stat.ML); Machine Learning (cs.LG); Methodology (stat.ME)
*备注: Accepted at the 43rd International Conference on Machine Learning (ICML 2026)
Abstract:This paper develops a general transfer learning framework for nonparametric regression with data consisting of multiple groups. Under the assumption that groups share a common structure along with group-specific deviations in additive form, the proposed method employs a two-stage offset learning procedure: the first stage pools data from all groups to estimate an overall mean function, and the second stage estimates offsets for each group, yielding final group-level estimators through additive combination. Upper bounds on the \mathcal L_2 error are established for the proposed framework, covering a broad class of nonparametric estimators under mild complexity and noise conditions. When instantiated with deep ReLU networks, explicit convergence rates are derived under hierarchical composition models, demonstrating the ability to overcome the curse of dimensionality. Conditions that enable positive transfer with faster rates are considered, including learning with simpler functions and data augmentation through pooling samples across groups. Various simulations and real-data experiments further validate the effectiveness of the proposed method.
[LG-54] Gravitational-wave parameter estimation with machine-learning generated surrogate waveforms
链接: https://arxiv.org/abs/2608.20222
作者: Suyog Garg,Kipp Cannon
类目: General Relativity and Quantum Cosmology (gr-qc); Instrumentation and Methods for Astrophysics (astro-ph.IM); Machine Learning (cs.LG)
*备注: 20 pp, 18 figs, under review in PRD
Abstract:The worldwide network of gravitational-wave detectors have detected more than 350 binary coalescence events till date. Future third-generation detectors, like Einstein telescope, are expected to detect orders-of-magnitude more signals from sources with more complicated characteristics, including eccentric orbits and high-mass ratio binaries. It is well-established that the computational cost of parameter estimation for signals from these kinds of sources will be extremely high. In particular, the process could be sped-up if generating theoretical waveform predictions, used for likelihood calculation becomes faster. Recently, various machine-learning techniques has been proposed to this end. In this work, we propose a two-stage deterministic conditional-autoencoder model for generating four-parameter SEOBNRv4 waveforms. The first-stage of the model generates amplitude and phase series of the waveform, while the second-stage calibrates the residual error in the predictions. Our model achieves a median mismatch of around 10^-2 with the target polarization waveforms, while the calibrated amplitude/phase series achieve 10^-6 level cosine distance error. We then propose a waveform conditioning step to enable use of these surrogate waveforms for downstream parameter estimation tasks. Finally, we perform extensive parameter estimation tests, with ML and EOB waveform injections and try to recover posterior estimates for the source parameters. We find that when ML waveforms are used to recover EOB target parameter estimates, the inferred posterior have some systematic bias. This inherent bias can be estimated and corrected for, and then importance reweighting of posterior samples can enable use of low-accuracy surrogate waveforms at low SNRs.
[LG-55] Discrete Diffusion Inference-Time Control with Nested Sequential Monte Carlo
链接: https://arxiv.org/abs/2608.20123
作者: Lohithsai Yadala Chanchu,Hany Abdulsamad,Christian A. Naesseth
类目: Machine Learning (stat.ML); Machine Learning (cs.LG)
*备注:
Abstract:We study inference-time control for text generation in discrete diffusion language models, where the goal is to steer sampling toward sequence-level rewards without retraining. Prior work in this domain has focused on particle-based methods such as best-of- n sampling and bootstrap sequential Monte Carlo, which may suffer from overoptimism and weight degeneracy, respectively. We address these limitations using \emphnested sequential Monte Carlo methods. We formulate nested SMC (NSMC) and fully-adapted nested SMC (FA-NSMC) for Feynman–Kac steering, identifying and correcting errors in prior formulations that lead to biased final estimates. We evaluate these methods on toxicity and fluency steering tasks, showing that NSMC and FA-NSMC consistently outperform best-of- n and bootstrap SMC.
[LG-56] Where Does the Union Bound Go? Best-Arm Identification and Strong FWER Control
链接: https://arxiv.org/abs/2608.19903
作者: Rianne de Heide
类目: Methodology (stat.ME); Machine Learning (cs.LG); Machine Learning (stat.ML)
*备注:
Abstract:In fixed-confidence best-arm identification, proofs often use a union bound across the competing arms. From a multiple-testing point of view this can look puzzling: if the best arm is unique, only one hypothesis of the form ``arm i is best’’ can be true. Why then should there be a Bonferroni-type factor of K-1 ? The answer is that there are two natural ways to orient the hypotheses. In one orientation, best-arm identification is literally a strong familywise-error-rate (FWER) problem with K-1 true nulls. In the opposite orientation, exactly one null is true, but a pairwise implementation can falsely reject that one null through any of K-1 comparisons. Thus the multiplicity has not disappeared; it just pops up in different places. This note makes the equivalence explicit in the terminology of both communities.
[LG-57] he impact of feature engineering and an optimisation framework for ocean colour machine learning
链接: https://arxiv.org/abs/2608.19899
作者: Edson Silva,Julien Brajard,Simon Cappe,Lasse H. Pettersson,François Counillon
类目: Atmospheric and Oceanic Physics (physics.ao-ph); Machine Learning (cs.LG)
*备注: Under review
Abstract:Machine learning (ML) is widely used for the development of ocean colour algorithms, but most studies focus on model parameter training and hyperparameter tuning. The optimisation of the data that feeds the models - i.e., Feature Engineering (FE) - is not fully explored. We assess the impact of FE in ocean colour machine learning models and we propose an optimisation framework that includes seven sequenced levels of data transformation: i. band choice, ii. log scaling, iii. spectral shape normalisation, iv. index extraction, v. principal component analysis, vi. feature scaling, and vii. zero-to-one scaling. We demonstrate the application for Multi-layer perceptron, Support Vector Machines, and eXtreme Gradient Boosting Trees on Sentinel-3 OLCI observations in the Norwegian coastal waters. The models are trained to estimate Chlorophyll-a concentration [Chl-a] and Secchi disk depth (Zsd). Results show that accuracy is highly variable among FE found in six studies using Sentinel-3 OLCI and the ones that we optimise. The R range from 0.01 to 0.55 for [Chl-a] and from 0.15 to 0.68 for Zsd, where the optimised FE shows the top results. The ML models with optimised FE could also improve by two times the R and reduce up to 63% of the mean absolute error when compared to CHL_OC4ME and CHL_NN standard algorithms. Nevertheless, no common optimised FE is found for all target variables and ML models, suggesting that FE optimisation is necessary for each application. Therefore, our proposed framework can be key for improving the accuracy of water quality monitoring in coastal waters.
[LG-58] Interpretable Feature Learning for RF Fingerprinting via Polar MKANs
链接: https://arxiv.org/abs/2608.19881
作者: Mikhail Krasnov,Ljupcho Milosheski,Carolina Fortuna
类目: ignal Processing (eess.SP); Machine Learning (cs.LG)
*备注: 5 pages, 5 figures, 3 tables
Abstract:Radio frequency (RF) fingerprinting authenticates wireless devices from hardware-induced I/Q impairments, typically with deep learning feature extractors that are accurate but opaque, limiting their use in security critical settings. We propose Polar Monotonic Kolmogorov-Arnold Networks (Polar MKAN), a block partitioned monotonic encoder on polar inputs in which each latent dimension depends exclusively on magnitude or phase, yielding channel separation and monotone responses by construction. On a synthetic gain and carrier frequency offset (CFO) benchmark, Polar MKAN reaches 57.2 percent DCI Disentanglement versus at most 12.9 percent for unpartitioned baselines. We further evaluate the detection accuracy trade off on real data and the sensitivity to blind CFO compensation.
[LG-59] A Repeated Measurements Approach to SoH Battery Modelling of Cyclic Aged Data in a Laboratory Environment
链接: https://arxiv.org/abs/2608.19879
作者: Mark Cary,Charles Bokor
类目: Methodology (stat.ME); Machine Learning (cs.LG); Systems and Control (eess.SY)
*备注: 17 pages, 9 figures, 2 tables
Abstract:This document describes the application of a first order linearised nonlinear repeated measurements approach to the analysis of battery cell ageing profiles generated under controlled conditions in a laboratory. The primary advantage of the model is it reflects the obvious structure in the data. Consequently, it is a two-component of variance model: variation within ageing profiles (measurement noise) and variation among ageing profiles (test-to-test or cell-to-cell) variation. Novel regularised iterative generalised least squares parameter identification schemes, with optimal hyper-parameter re-estimation, are used to identify the hierarchical nonlinear model. The training data comprised SoH profiles for 10 cells aged at various constant discharge and charge current cycles at a fixed chamber environmental temperature of 25 [ ^\circ C]. Each cell SoH profile is modelled using a simple power law expression, whereas the variation in ageing parameters is modelled using a single knot cubic B-spline. SoH is accurately predicted to \pm 0.191% for SOH \in [0,20] .
[LG-60] Learning piecewise-smooth dynamical systems
链接: https://arxiv.org/abs/2608.19785
作者: Davide Murari,Erik Jansson,Chris Budd OBE,Carola-Bibiane Schönlieb
类目: Dynamical Systems (math.DS); Machine Learning (cs.LG); Numerical Analysis (math.NA)
*备注:
Abstract:Discovering dynamical systems from trajectory data is a central problem in applied mathematics and engineering. Whilst recent advances in machine learning have led to strong progress in data-driven system identification, much less attention has been given to systems with discontinuous dynamics. These systems are nevertheless highly relevant in applications, including climate dynamics and mechanical systems with friction. In this work, we consider the problem of identifying piecewise-smooth dynamical systems directly from trajectory data. Compared with the smooth setting, this requires recovering the governing equations and detecting the switching hyperplanes that separate different dynamical regimes and characterising their behaviour, such as sliding motion. We present a modular framework for discovering such systems by first estimating switching hyperplanes from data and then learning smooth dynamics within each region using geometry-constrained neural networks. The geometry-learning phase is studied from a statistical perspective, analysing the identifiability of the discontinuities and the robustness of the procedure. We also introduce a novel neural network architecture with a prescribed discontinuity set, and provide a theoretical analysis of its approximation properties. The approach is tested on low-dimensional benchmark problems, including dry-friction oscillators and the PP04 climate model for the ice ages.
[LG-61] skchange: Fast and Flexible Algorithms for Changepoint Detection
链接: https://arxiv.org/abs/2608.19767
作者: Martin Tveten,Johannes Voll Kolstø,Per August Jarval Moen
类目: Computation (stat.CO); Machine Learning (cs.LG)
*备注: 6 pages, 3 figures
Abstract:Skchange is an open-source Python library for detecting structural changes in time series. It implements modern change detection algorithms within a unified and extensible framework. The algorithms are modular and composable, and they include changepoint search methods based on both cost minimisation and statistical tests. Key features include the detection of anomalous segments in addition to changepoints; theoretically well-founded fast and approximate search methods; theoretically well-founded algorithms for high-dimensional data, covering settings where either few or many features change simultaneously; utilities for automatic and data-driven penalty calibration, which balances false alarms against missed detections; and a large collection of built-in costs and statistical tests. The design follows established scikit-learn conventions to streamline both user and contributor experience, and Numba is used extensively to achieve high computational performance. Source code and documentation are available at this https URL.
[LG-62] Composition-Driven Phase Evolution in Sm-Doped BiFeO3 via Latent-Field Reconstruction of Atomically Resolved STEM Data
链接: https://arxiv.org/abs/2608.19544
作者: Newsha Javanmardi,Christopher T. Nelson,Anna N. Morozovska,Eugene A. Eliseev,Ichiro Takeuchi,Sergei V. Kalinin
类目: Materials Science (cond-mat.mtrl-sci); Machine Learning (cs.LG)
*备注:
Abstract:Functionalities of ferroelectric materials are governed by the spatial organization and coupling of polarization, strain, lattice rotation, and structural order accessible via atomically resolved scanning transmission electron microscopy (STEM) images. Quantitative interpretation of atomic-resolution STEM data has conventionally relied on locating atomic columns and converting their fitted coordinates into local structural descriptors. Here, we develop a field-based approach in which atomic-resolution images are represented by spatially varying latent Bragg fields, whose amplitudes and phases provide continuous maps of crystalline order, lattice displacement, strain, rotation, and mode-specific residual structure. The observed atomically resolved images are decoded from the latent fields. We apply this framework to image series of Sm-substituted BiFeO3 spanning 0-20% Sm and crossing the composition-driven boundary between the R3c ferroelectric phase and the orthorhombic, nonpolar Pnma phase. Conventional atom-resolved parameterization is used as an independent validation, showing that reconstructed Bragg amplitude tracks local atomic-column intensity and that field-derived shear reproduces unit-cell angular distortions obtained from atom fitting. The combined analysis reveals a systematic evolution from extended ferroelectric domains at low Sm concentration, through the appearance and growth of localized regions with period-doubled Pnma order at intermediate compositions, to a connected Pnma-dominated state at high Sm content. The period-doubled order is accompanied by enhanced shear and lattice rotation and by progressive reorganization of the ferroelectric domain structure. These results establish latent-field reconstruction as a physically interpretable complement to atom finding and provide a unified framework for resolving composition-driven phase evolution in ferroic materials.
[LG-63] Microlensify: a Transformer Based Machine Learning Classifier for Microlensing Events Trained on TESS Light Curves
链接: https://arxiv.org/abs/2608.19419
作者: Atousa Kalantari,Somayeh Khakpash,Sedighe Sajadian,Hosein Haghi,Willow Fox Fortino,Rosanne Di Stefano
类目: Instrumentation and Methods for Astrophysics (astro-ph.IM); Earth and Planetary Astrophysics (astro-ph.EP); Astrophysics of Galaxies (astro-ph.GA); Solar and Stellar Astrophysics (astro-ph.SR); Machine Learning (cs.LG)
*备注:
Abstract:Microlensing can reveal populations of faint compact objects that are otherwise difficult to detect. Depending on their design, all-sky surveys have the potential to search for these objects across the sky. The Transiting Exoplanet Survey Satellite (TESS), primarily designed to detect transiting exoplanets, also provides near all-sky coverage with high cadence. In this work, we use TESS data to search for microlensing candidates using both traditional and machine-learning methods and to identify associated false positives in high-cadence surveys. Microlensify is a physics-informed, transformer-based variational autoencoder trained on simulated single-lens microlensing light curves and real TESS Sector 12 data. The model classifies events, reconstructs light curves, and estimates microlensing event durations. Applied to \sim 5.6 million TESS light curves, it identified between 0.036% and 1.89% as microlensing candidates across different TESS pipelines. After applying microlensing detection metrics and cross-matching with SIMBAD, we obtained a final list of candidates and identified false positives including long-period variables, Mira variables, cataclysmic variables, red giants, and transients. We also found Gaussian-like peaks caused by asteroid crossings, a potential source of false positives in high-cadence microlensing surveys. The model also predicts event duration with an accuracy of R^2 = 0.97 . The model was further tested on published events from different ground-based microlensing surveys, confirming 92.7% as microlensing, demonstrating its applicability across surveys with different cadences. Subjects: Instrumentation and Methods for Astrophysics (astro-ph.IM); Earth and Planetary Astrophysics (astro-ph.EP); Astrophysics of Galaxies (astro-ph.GA); Solar and Stellar Astrophysics (astro-ph.SR); Machine Learning (cs.LG) Cite as: arXiv:2608.19419 [astro-ph.IM] (or arXiv:2608.19419v1 [astro-ph.IM] for this version) https://doi.org/10.48550/arXiv.2608.19419 Focus to learn more arXiv-issued DOI via DataCite (pending registration) Submission history From: Atousa Kalantari [view email] [v1] Wed, 19 Aug 2026 20:11:01 UTC (4,248 KB)
[LG-64] Deep-MKV-TS: Path-Dependent McKean–Vlasov Control for Financial Time Series Generation
链接: https://arxiv.org/abs/2608.19394
作者: Samer El Boustany,Théo Basseras,Samy Mekkaoui,Alexandre Alouadi,Yadh Hafsi,Huyên Pham
类目: Computational Finance (q-fin.CP); Computational Engineering, Finance, and Science (cs.CE); Machine Learning (cs.LG); Optimization and Control (math.OC)
*备注:
Abstract:We introduce Deep-MKV-TS, a path-dependent McKean-Vlasov framework for financial scenario generation. The stochastic dynamics are chosen by matching selected path and volatility features of generated scenarios to those observed in the data. Starting from an interpretable reference model, Deep-MKV-TS preserves the reference drift and adjusts its volatility, while a regularization penalty limits unnecessary departures from the calibrated dynamics. We solve the resulting control problem using a neural, sample-based implementation of the stochastic maximum principle. We validate the method against an exactly computable oracle. On Heston and Heston-mixture models, Deep-MKV-TS substantially reduces path-dependent and volatility-related deficiencies of the reference model. In delayed-volatility experiments, the correction remains effective as the forecasting horizon increases, while direct training becomes less reliable. On held-out intraday equity-index futures, the corrected model improves conditional forecasts relative to the reference and reaches a level of performance comparable to flexible generative and historical baselines. The resulting scenarios also support greater exposure than the reference under a fixed drawdown-risk target. These results show that path-dependent McKean-Vlasov control can enrich an interpretable reference model without replacing it. Subjects: Computational Finance (q-fin.CP); Computational Engineering, Finance, and Science (cs.CE); Machine Learning (cs.LG); Optimization and Control (math.OC) Cite as: arXiv:2608.19394 [q-fin.CP] (or arXiv:2608.19394v1 [q-fin.CP] for this version) https://doi.org/10.48550/arXiv.2608.19394 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
[LG-65] Heteroscedastic Neural Surrogate Modeling for Robust and Rapid Bayesian Inference in Fusion Plasma Diagnostics
链接: https://arxiv.org/abs/2608.19377
作者: Liyun Zhang,Naoya Mamada,Kentaro Sakai,Takeo Hoshi,Toru Aonishi
类目: Computation (stat.CO); Machine Learning (cs.LG)
*备注: 8 pages, 3 figures
Abstract:Bayesian inference via Markov Chain Monte Carlo (MCMC) provides effective parameter estimation, but its real-time application in complex physical systems is hindered by heavy computational bottlenecks and extreme sensitivity to statistical noise. We address this by proposing a neural-network-based probabilistic surrogate framework for rapid and robust MCMC inference. Using fusion plasma Thomson scattering diagnostics as a challenging, noise-dominated testbed, our approach employs a dual-head architecture to simultaneously estimate the expected physical emission spectrum and the channel-wise intrinsic measurement noise variance. By optimizing a Gaussian Negative Log-Likelihood (GNLL) objective, the learned aleatoric uncertainty dynamically buffers the sampler against pathological shot noise. Evaluations demonstrate that this surrogate framework achieves 1500x acceleration over exact physical forward models, while simultaneously reducing inference error (RMSE) by 20% compared to standard homoscedastic neural baselines, offering a highly promising paradigm for real-time physical analysis.
[LG-66] Deep neural networks as lattice gauge theories
链接: https://arxiv.org/abs/2608.19331
作者: Ro Jefferson,Shradha Ramakrishnan
类目: High Energy Physics - Theory (hep-th); Disordered Systems and Neural Networks (cond-mat.dis-nn); Machine Learning (cs.LG)
*备注: 40 pages, infinite figures
Abstract:We modify the NN/QFT duality [1] to incorporate the layerwise permutation symmetry of the network, resulting in a (0!+!1) -dimensional lattice gauge theory, in which each layer of N neurons acts as an N -component lattice site, and the weight matrices play the role of gauge fields living on the links. In this framework, we compute the tree-level neuron-neuron propagator which describes the evolution of layer variance in the network, and develop the Feynman diagram machinery to compute interactions in the perturbative expansion in 1/N . In particular, we obtain a recursive expression for all corrections to the exact propagator at O(1) , representing statistical fluctuations in the ensemble of networks, including infinitely-many loop diagrams mediating the interactions from previous layers. We also present a preliminary analysis of neuron scattering amplitudes that contribute order-by-order in 1/N , which provides a field-theoretic framework for studying higher-point correlations, and by extension information propagation, in deep networks. We remark on some interesting directions for future work at the intersection of neural networks and quantum field theory.
[LG-67] Quantum Gaussian processes for prediction of channel observations
链接: https://arxiv.org/abs/2608.19306
作者: Jonas Jäger,Yaroslav Khmelnitskiy,Paolo Braccia,Artur Miroszewski,Diego García-Martín,M. Cerezo,Piotr Czarnik
类目: Quantum Physics (quant-ph); Machine Learning (cs.LG); Machine Learning (stat.ML)
*备注: 14 + 7 pages, 5 + 2 figures
Abstract:Given a set of input states, we consider the task of predicting the expectation value of a Pauli observable at the output of an unknown quantum evolution, using only a limited number of measurements. Recently, quantum Gaussian process (QGP) regression was introduced for this task across various classes of unitary evolution. Here, we extend the QGP framework beyond unitary dynamics. In particular, we prove convergence of the channel’s outputs to a QGP and derive the associated closed-form kernel under a uniform (Lebesgue measure) prior over quantum channels. The kernel’s dimensional factor, however, dictates the required observation precision. While manageable when the channel and observable are restricted to small subsystems, exponential suppression precludes learning when the subsystem grows extensively with the system size. Since the Lebesgue prior is overly broad for many applications, we propose an empirical Bayes heuristic that replaces the dimensional factor with a learnable scale parameter while retaining the kernel’s state-overlap correlation structure. In numerical simulations of up to 64 qubits, channel QGP regression with the Lebesgue kernel exhibits a strong inductive bias for local channels, enabling faithful extrapolation. For global 64-qubit channels, the rescaled kernel restores learnability, with predictions improving systematically with the shot budget. Results from a noisy quantum computer further demonstrate the robustness of QGP regression under experimental conditions. Beyond regression, we validate QGPs as Bayesian-optimization surrogates for state preparation under noisy XXZ dynamics.
[LG-68] Recovering Nonlinear Functions of Latent Variables: A Plausible-Value Neural Network Framework
链接: https://arxiv.org/abs/2608.19282
作者: Eunjeong Song(1),Sehee Hong(1) ((1) Department of Education, Korea University, Seoul, Republic of Korea)
类目: Methodology (stat.ME); Machine Learning (cs.LG)
*备注: 51 pages, 4 figures, 24 tables. Data, materials, and code: this https URL
Abstract:When factor scores replace true latent scores in nonlinear prediction, measurement error attenuates the recoverable variance of any k th-order component of the regression function by \rho^k – the k th power of the score’s coefficient of determination – for any linear score type. This study derives the bound via Hermite polynomial expansion and proposes PV-ANN – plausible values (posterior draws preserving latent variance) combined with artificial neural networks (learning functional form without prespecification). The bound governs recovery of the latent-scale function, not prediction of the outcome from observed indicators, for which factor scores are already sufficient; the two metrics are therefore predicted to dissociate. An 18-condition simulation supports both predictions: in the nonlinear low-reliability conditions PV-ANN closes about four fifths of the function-shape recovery gap between a factor-score learner and one given the true latent values, and the margin widens as reliability falls, while predictive accuracy is not improved, as the theory requires. A Big Five application illustrates the intended exploratory workflow and delineates boundary conditions under weak signal and measurement model misspecification.
[LG-69] Demons on a Budget: Adaptive Measurement Placement at the Entanglement Phase Transition
链接: https://arxiv.org/abs/2608.19248
作者: Rohan Pandey
类目: Quantum Physics (quant-ph); Machine Learning (cs.LG)
*备注: 10 pages
Abstract:Monitored quantum circuits exhibit a measurement-induced phase transition between volume-law and area-law entanglement as a function of the measurement rate p . Prior work places measurements at random locations and treats the rate as the control parameter. We instead fix the measurement budget and vary the placement process, comparing random placement against hand-designed and learned policies in brickwork random Clifford circuits at matched budget. First, placement geometry matters more than placement information. A deterministic contiguous sweep cuts the half-cut entropy by a factor of 3.4 relative to random placement, while equal-coverage unstructured placement and a greedy policy with full state access do far worse. The effect is carried by spatial order alone: measuring the k least recently measured sites gives 4.14 \pm 0.06 bits with random tie-breaking and 1.29 \pm 0.04 bits with position-ordered tie-breaking. Second, the sweep eliminates the transition rather than shifting it. Tripartite mutual information crossings recede as p^* \propto 1/L , the steady-state entropy saturates at an L -independent ceiling near 0.46/p , and data for 64 \le L \le 512 collapse onto the form S = p^-1 f(pL) predicted by a ballistic regrowth argument. Third, in stabilizer dynamics every outcome is deterministic or a fair coin flip, so the record’s Shannon entropy is exactly countable; the sweep dominates the entropy-versus-record-cost frontier while paying the same roughly one bit per measurement as random placement. Policies trained by cross-entropy and proximal policy optimization do not find the sweep: score-based policies parameterize which sites to measure, not the order in which degenerate scores are resolved, and the effect lives in that order. The phase diagram of monitored dynamics is a property of the placement process, not only of the measurement rate.
[LG-70] orchDCM: A Unified PyTorch-Native Package for Discrete Choice Modeling
链接: https://arxiv.org/abs/2608.19231
作者: Baichuan Mo,Zhengzhong Ricky You,Xiqun Michael Chen,Ruimin Li
类目: Machine Learning (stat.ML); Machine Learning (cs.LG)
*备注:
Abstract:Estimating large and simulation-intensive discrete choice models (DCMs) requires repeated evaluation of utilities, probabilities, derivatives, and simulated likelihoods over many observations, alternatives, and draws. Existing DCM software provides mature econometric workflows, while recent GPU-oriented tools accelerate selected models, leaving a gap between econometric coverage and scalable differentiable computation. We introduce TorchDCM, an open Python package for discrete choice modeling that compiles choice data and model specifications into a unified PyTorch-native likelihood engine for estimation, inference, prediction, and structured reporting on CPU or CUDA devices. The package covers the principal econometric functionality available across Biogeme and Apollo, including multinomial, nested, mixed, ordered, latent-variable, and panel likelihoods. It also supports ragged choice sets, constrained parameters, covariance estimation, willingness-to-pay analysis, elasticities, and extensible likelihood components. We evaluate TorchDCM against seven other estimation packages in aligned synthetic and real-data full-estimation experiments. TorchDCM completes all 45 synthetic cases, runs fastest in every comparable synthetic case, and satisfies the prespecified final-log-likelihood tolerance in every comparison with at least two comparable solutions. More precisely, it reduces median runtime by 89.1%-99.7% relative to Biogeme and Apollo across model-data settings. CUDA provides an additional 12.0-71.0x speedup over single-core TorchDCM. These results establish a scalable and reproducible foundation for econometric estimation and differentiable choice-model development. The open-source package and executed examples are available at this https URL.
[LG-71] M3: A State-Event Generative Foundation Model for Market Microstructure Dynamics
链接: https://arxiv.org/abs/2608.19227
作者: Yanzhi Zhang,Yu Ma,Yilin Cheng,Jian Li,Yitong Duan
类目: Computational Finance (q-fin.CP); Machine Learning (cs.LG)
*备注:
Abstract:Market microstructure simulation aims to model how liquidity, prices, and order flow evolve in electronic financial markets. Since market data reveal only one realized trajectory, many important questions are inherently counterfactual and require realistic trajectory-level simulation. Existing financial generative models, however, often model order events and market states, such as the LOB, in isolation, overlooking the dynamic interaction between order flow and liquidity in market microstructure. We propose the \textbfM3 (\underlineMarket \underlineMicrostructure \underlineModel), a state-event generative foundation model for market microstructure dynamics. \textbfM3 learns to generate future order-flow trajectories, while accounting for the evolving interaction between order events and limit-order-book liquidity. Trained on large-scale order-level real stock market data, \textbfM3 exhibits predictable scaling behavior, reproduces key market stylized facts, and enables practical simulation-based applications including forecasting, stress testing, and market-impact analysis. These results suggest a scalable foundation-model paradigm for counterfactual market simulation at the microstructure level.
附件下载
