cs.AI · 2026-05-22 · No. 7

Artificial Intelligence, 2026-05-22.

68 new papers in cs.AI. Titles, authors, abstracts. Links to arXiv. Want this in your inbox every morning? Subscribe →

01 — The papers

68 entries
  1. 01

    MOSS: Self-Evolution through Source-Level Rewriting in Autonomous Agent Systems

    Qianshu Cai, Yonggang Zhang, Xianzhang Jia, Wei Xue, Jun Song, Xinmei Tian, Yike Guo

    cs.AI · cs.LG

    Autonomous agentic systems are largely static after deployment: they do not learn from user interactions, and recurring failures persist until the next human-driven update ships a fix. Self-evolving agents have emerged in response, but all confine evolution to text-mutable artifacts -- skill files, prompt configurations, memory schemas, workflow graphs -- and leave the agent harness untouched. Since routing, hook ordering, state invariants,...

    arxiv.org/abs/2605.22794 · PDF

  2. 02

    Gated DeltaNet-2: Decoupling Erase and Write in Linear Attention

    Ali Hatamizadeh, Yejin Choi, Jan Kautz

    cs.AI

    Linear attention replaces the unbounded cache of softmax attention with a fixed-size recurrent state, reducing sequence mixing to linear time and decoding to constant memory. The hard part is not just what to forget, but how to edit this compressed memory without scrambling existing associations. Delta-rule models subtract the current read before writing a new value, and Kimi Delta Attention (KDA) sharpens forgetting with channel-wise decay....

    arxiv.org/abs/2605.22791 · PDF

  3. 03

    LCGuard: Latent Communication Guard for Safe KV Sharing in Multi-Agent Systems

    Sadia Asif, Mohammad Mohammadi Amiri, Momin Abbas, Prasanna Sattigeri, Karthikeyan Natesan Ramamurthy

    cs.AI · cs.ET · cs.LG · cs.MA

    Large language model (LLM)-based multi-agent systems increasingly rely on intermediate communication to coordinate complex tasks. While most existing systems communicate through natural language, recent work shows that latent communication, particularly through transformer key-value (KV) caches, can improve efficiency and preserve richer task-relevant information. However, KV caches also encode contextual inputs, intermediate reasoning...

    arxiv.org/abs/2605.22786 · PDF

  4. 04

    Deep Reinforcement Learning for Flexible Job Shop Scheduling with Random Job Arrivals

    Yu Tang, Muhammad Zakwan, Efe Balta, John Lygeros, Alisa Rupenyan

    cs.AI · math.OC

    The Flexible Job Shop Scheduling Problem (FJSP) is the optimal allocation of a set of jobs to machines. Two primary challenges persist in FJSP: the unpredictable arrival of future jobs and the combinatorial complexity of the problem, rendering it intractable for conventional mixed-integer linear programming solvers. This paper proposes an event-based \gls{DRL} approach to solve FJSP with random job arrivals. Specifically, we employ the...

    arxiv.org/abs/2605.22773 · PDF

  5. 05

    Advancing Mathematics Research with AI-Driven Formal Proof Search

    George Tsoukalas, Anton Kovsharov, Sergey Shirobokov, Anja Surina, Moritz Firsching, Gergely Bérczi, Francisco J. R....

    cs.AI

    Large language models (LLMs) increasingly excel at mathematical reasoning, but their unreliability limits their utility in mathematics research. A mitigation is using LLMs to generate formal proofs in languages like Lean. We perform the first large-scale evaluation of this method's ability to solve open problems. Our most capable agent autonomously resolved 9 of 353 open Erdős problems at the per-problem cost of a few hundred dollars, proved...

    arxiv.org/abs/2605.22763 · PDF

  6. 06

    Towards a General Intelligence and Interface for Wearable Health Data

    Girish Narayanswamy, Maxwell A. Xu, A. Ali Heydari, Samy Abdel-Ghaffar, Marius Guerard, Kara Vaillancourt, Zhihan...

    cs.AI

    While ubiquitous wearable sensors capture a wealth of behavioral and physiological information, effectively transforming these signals into personalized health insights is challenging. Specifically, converting low-level sensor data into representations capable of characterizing higher-level states is difficult due to high phenotypic diversity and variation in individual baseline health, physiology, and lifestyle factors. Moreover, collecting...

    arxiv.org/abs/2605.22759 · PDF

  7. 07

    HarnessAPI: A Skill-First Framework for Unified Streaming APIs and MCP Tools

    Edwin Jose

    cs.AI · cs.SE

    Every Python function deployed as an LLM tool must today exist in two forms: an HTTP endpoint for human-facing clients and CI pipelines, and an MCP tool registration for agent runtimes such as Claude and Cursor. These representations share business logic yet diverge in all the surrounding machinery (routing, validation, serialisation, streaming, and schema maintenance), and they drift apart as the underlying code evolves. We present...

    arxiv.org/abs/2605.22733 · PDF

  8. 08

    Beyond Acoustic Emotion Recognition: Multimodal Pathos Analysis in Political Speech Using LLM-Based and Acoustic Emotion Models

    Juergen Dietrich

    cs.AI · cs.CL · cs.HC · cs.SD · eess.AS

    We investigate whether acoustic emotion recognition models can serve as proxies for the Pathos dimension in political speech analysis, as operationalised by the TRUST multi-agent large language model (LLM) pipeline. Using a Bundestag plenary speech by Felix Banaszak (51 segments, 245 s) as a case study, we compare three analysis modalities: (1) emotion2vec_plus_large, an acoustic speech emotion recognition (SER) model whose continuous Arousal...

    arxiv.org/abs/2605.22732 · PDF

  9. 09

    Can AI Make Conflicts Worse? An Alignment Failure in LLM Deployment Across Conflict Contexts

    Andrii Kryshtal

    cs.AI · cs.HC

    AI models are already deployed in societies affected by armed conflict, and journalists, humanitarian workers, governments and ordinary citizens rely on them for information or for their work processes. No established practice exists for checking whether their outputs can make those conflicts worse. We tested nine model configurations from four providers (OpenAI, Anthropic, DeepSeek, xAI) on 90 multi-turn scenarios designed to surface...

    arxiv.org/abs/2605.22720 · PDF

  10. 10

    Parametric Modular Answer Set Programs Made Declarative

    Jorge Fandinno, Yuliya Lierler, Torsten Schaub

    cs.AI · cs.LO

    In this paper, we explore the concept of modularity in first-order answer set programming (ASP). We introduce a new formalism called parametric modular logic programs, which allows defining subprograms with parameters and intensionality statements. We demonstrate how this formalism can capture the semantics of clingo-programs with collective control, a feature that enables structuring and instantiating subprograms. We provide theoretical...

    arxiv.org/abs/2605.22716 · PDF

  11. 11

    AMEL: Accumulated Message Effects on LLM Judgments

    Sid-ali Temkit

    cs.AI · cs.CL · cs.LG

    Large language models are routinely used as automated evaluators: to review code, moderate content, or score outputs, often with many items passing through one conversation. We ask whether the polarity of prior conversation history biases subsequent judgments, an effect we call the accumulated message effect on LLM judgments (AMEL). Across 75,898 API calls to 11 models from 4 providers (OpenAI, Anthropic, Google, and four open-source models),...

    arxiv.org/abs/2605.22714 · PDF

  12. 12

    Beyond the Org Chart: AI and the Transformation of Invisible Work

    Stephanie Rosenthal, Shamsi Iqbal

    cs.AI · cs.HC

    An increasing number of news and research articles report that AI adoption is allowing professionals to blur and extend the boundaries of their corporate roles. With the goal of understanding how work processes might be changing in an AI-forward company, we interviewed 24 product-focused individuals at a large technology firm about how AI has impacted their own work, their work within their product team, and their professional interactions....

    arxiv.org/abs/2605.22707 · PDF

  13. 13

    Forecasting Scientific Progress with Artificial Intelligence

    Sean Wu, Pan Lu, Yupeng Chen, Jonathan Bragg, Yutaro Yamada, Peter Clark, David Clifton, Philip Torr, James Zou, Junchi Yu

    cs.AI

    Artificial intelligence (AI) is increasingly embedded in scientific discovery, yet whether it can anticipate scientific progress remains unclear. To study this question, we introduce a temporally grounded evaluation framework for forecasting scientific progress under controlled knowledge constraints. We present CUSP (Cutoff-conditioned Unseen Scientific Progress), a multi-disciplinary and event-level benchmark that evaluates scientific...

    arxiv.org/abs/2605.22681 · PDF

  14. 14

    Is Capability a Liability? More Capable Language Models Make Worse Forecasts When It Matters Most

    Nick Merrill, Jaeho Lee, Ezra Karger

    cs.AI

    We document inverse scaling in LLMs on forecasting problems whose underlying time series exhibit superlinear growth and tail risk of regime change, a structure common in finance and epidemiology. On these tasks, more capable models produce worse distributional forecasts. The pattern appears on ForecastBench-Sim (FBSim), a contamination-free, simulated-world benchmark we release, in forecasting synthetic SIR epidemics with a matched linear...

    arxiv.org/abs/2605.22672 · PDF

  15. 15

    WorkstreamBench: Evaluating LLM Agents on End-to-End Spreadsheet Tasks in Finance

    Thomson Yen, Julian Poeltl, Harshith Srinivas Gear, Yilin Meng, Joshua Fan, Adam Shen, Yili Liu, Ali Bauyrzhan, Siri...

    cs.AI

    LLM agents are increasingly expected to carry out end-to-end workflows, producing complete artifacts from high-level user instructions. To meet enterprise needs, frontier AI labs have developed agents that can construct entire spreadsheets from scratch. This is especially relevant in finance, where core workflows such as financial modeling, forecasting, and scenario analysis are commonly conducted through spreadsheets. Yet, existing...

    arxiv.org/abs/2605.22664 · PDF

  16. 16

    Claw AI Lab: An Autonomous Multi-Agent Research Team

    Fan Wu, Cheng Chen, Zhenshan Tan, Taiyu Zhang, Xinzhen Xu, Yanyu Qian, Dingcheng Gao, Lanyun Zhu, Qi Zhu, Yi Tan,...

    cs.AI

    We present Claw AI Lab, a lab-native autonomous research platform that advances automated research from a hidden prompt-to-paper pipeline into an interactive AI laboratory. Rather than centering the system around a single agent or a fixed serial workflow, we allow users to instantiate a full research team from one prompt, with customizable roles, collaborative workflows, real-time monitoring, artifact inspection, and rollback/resume control...

    arxiv.org/abs/2605.22662 · PDF

  17. 17

    AtelierEval: Agentic Evaluation of Humans & LLMs as Text-to-Image Prompters

    Hanjun Luo, Zhimu Huang, Sylvia Chung, Yiran Wang, Yingbin Jin, Jialin Li, Jiang Li, Xinfeng Li, Hanan Salam

    cs.AI

    Text-to-image (T2I) systems increasingly rely on upstream prompters, either humans or multimodal large language models (MLLMs), to translate user intent into detailed prompts. Yet current benchmarks fix the prompt and only evaluate T2I models, leaving the prompting proficiency of this upstream component entirely unmeasured. We introduce AtelierEval, the first unified benchmark that quantifies prompting proficiency across 360 expert-crafted...

    arxiv.org/abs/2605.22645 · PDF

  18. 18

    Spreadsheet-RL: Advancing Large Language Model Agents on Realistic Spreadsheet Tasks via Reinforcement Learning

    Banghao Chi, Yining Xie, Mingyuan Wu, Jingcheng Yang, Jize Jiang, Zhaoheng Li, Shengyi Qian, Minjia Zhang, Klara...

    cs.AI

    Spreadsheet systems (e.g., Microsoft Excel, Google Sheets) play a central role in modern data-centric workflows. As AI agents grow increasingly capable of automating complex tasks, such as controlling computers and generating presentations, building an AI-driven spreadsheet agent has emerged as a promising research direction. Most existing spreadsheet agents rely on specialized prompting over general-purpose LLMs; while this design has...

    arxiv.org/abs/2605.22642 · PDF

  19. 19

    Think Thrice Before You Speak: Dual knowledge-enhanced Theory-of-Mind Reasoning for Persuasive Agents

    Minghui Ma, Bin Guo, Runze Yang, Mengqi Chen, Yan Liu, Jingqi Liu, Yahan Pei, Xuehao Ma, Qiuyun Zhang, Zhiwen Yu

    cs.AI

    Persuasive dialogue requires reasoning about others' latent mental states, a capability known as Theory of Mind (ToM). However, due to reliance on simple prompting strategies and insufficient ToM knowledge, existing LLMs often fail to capture the intrinsic dependencies among mental states, leading to fragmented representations and unstable reasoning. To address these challenges, we introduce the ToM-based Persuasive Dialogue (ToM-PD) task,...

    arxiv.org/abs/2605.22602 · PDF

  20. 20

    TerminalWorld: Benchmarking Agents on Real-World Terminal Tasks

    Zhaoyang Chu, Jiarui Hu, Xingyu Jiang, Pengyu Zou, Han Li, Chao Peng, Peter O'Hearn, Earl T. Barr, Mark Harman,...

    cs.AI

    We introduce TerminalWorld, a scalable data engine that automatically reverse-engineers high-fidelity evaluation tasks from "in-the-wild" terminal recordings. Processing 80,870 terminal recordings, the engine yields a full benchmark of 1,530 validated tasks, spanning 18 real-world categories, ranging from short everyday operations to workflows exceeding 50 steps, and covering 1,280 unique commands. From these, we curate a Verified subset of...

    arxiv.org/abs/2605.22535 · PDF

  21. 21

    A Subjective Logic-based method for runtime confidence updates in safety arguments

    Benjamin Herd, Jessica Kelly, Clarissa Heinemann, João-Vitor Zacchi

    cs.AI

    We present a method for dynamic quantitative assurance that enhances static safety cases with continuous, runtime-driven confidence updates. The method quantifies and propagates confidence across the development lifecycle by integrating design-time evidence and windowed runtime Safety Performance Indicators (SPIs) within a single Subjective Logic (SL)-based assurance case. At runtime, SPI evidence is continuously evaluated, and targeted...

    arxiv.org/abs/2605.22530 · PDF

  22. 22

    Meta-Learning for Rapid Adaptation in Reference Tracking of Uncertain Nonlinear Systems

    Jiaqi Yan, Ankush Chakrabarty, Niklas Schmid, John Lygeros, Alisa Rupenyan

    cs.AI

    In this paper, we address the problem of reference tracking for uncertain nonlinear systems. Since collecting data from the target system (i.e., the system of interest) is often challenging, our objective is to design optimal controllers using limited target system data. Meta-learning provides a promising paradigm by leveraging offline data from source systems (systems sharing structural similarities with the target system) to accelerate...

    arxiv.org/abs/2605.22513 · PDF

  23. 23

    Search-E1: Self-Distillation Drives Self-Evolution in Search-Augmented Reasoning

    Zihan Liang, Yufei Ma, Ben Chen, Zhipeng Qian, Xuxin Zhang, Huangyu Dai, Lingtao Mao

    cs.AI · cs.CL · cs.IR

    Post-training has become the dominant recipe for turning a language model into a competent search-augmented reasoning agent. A line of recent work pushes its performance further by adding elaborate machinery on top of this standard pipeline. These augmentations import external supervision from stronger external systems, attach auxiliary modules such as process reward models or retrospective critics, restructure the rollout itself with tree...

    arxiv.org/abs/2605.22511 · PDF

  24. 24

    Towards Direct Evaluation of Harness Optimizers via Priority Ranking

    Kai Tzu-iunn Ong, Minseok Kang, Dongwook Choi, Junhee Cho, Seungju Kim, Seungwon Lim, Geunha Jang, Minwoo Oh,...

    cs.AI

    Harness optimization enables automated agent creation by having an optimizer agent iteratively update the harness of target agents. Despite its success, current studies evaluate optimizers solely by observing target agents' performance gains. This indirect end-improvement evaluation neglects optimizers' actions at intermediate steps, which are often erroneous and hinder agent performance. Therefore, it is unclear whether harness optimization...

    arxiv.org/abs/2605.22505 · PDF

  25. 25

    LACO: Adaptive Latent Communication for Collaborative Driving

    Tianhao Chen, Yuheng Wu, Dongman Lee

    cs.AI · cs.CV

    Collaborative driving aims to improve safety and efficiency by enabling connected vehicles to coordinate under partial observability. Recent approaches have evolved from sharing visual features for perception to exchanging language-based reasoning through foundation models for behavioral coordination. Though communicating in language provides intuitive information, it introduces two challenges: high latency caused by autoregressive decoding...

    arxiv.org/abs/2605.22504 · PDF

  26. 26

    Compiling Agentic Workflows into LLM Weights: Near-Frontier Quality at Two Orders of Magnitude Less Cost

    Simon Dennis, Rivaan Patil, Kevin Shabahang, Hao Guo

    cs.AI · cs.LG

    Agent orchestration frameworks have proliferated, collectively exceeding 290,000 GitHub stars across LangGraph, CrewAI, Google ADK, OpenAI Agents SDK, Semantic Kernel, Strands, and LlamaIndex. All follow the same pattern: an external orchestrator above the LLM, injecting instructions and routing decisions every turn. Recent work has shown this architecture is dominated for procedural tasks by simply providing the procedure in a frontier...

    arxiv.org/abs/2605.22502 · PDF

  27. 27

    KAPPS: A knowledge-based CPPS Architecture for the Circular Factory

    Etienne Hoffmann, Jan-Felix Klein, Sören Weindel, Max Goebels, Sebastian Behrendt, Daniel Hernández, Ratan Bahadur...

    cs.AI · eess.SY

    While linear manufacturing relies on homogeneous materials and predefined process sequences, circular manufacturing reintroduces used products with heterogeneous and uncertain conditions. This shift demands manufacturing systems capable of handling variable product states, dynamically reconfigurable processes, and the integration of human and machine knowledge. Conventional manufacturing IT architectures, designed for stable structures and...

    arxiv.org/abs/2605.22457 · PDF

  28. 28

    S2ED: From Story to Executable Descriptions for Consistency-Aware Story Illustration

    Sijing Yin, Jiamou Liu, Xiao Tang, Yaser Shakib, Qian Liu

    cs.AI

    Multi-frame story illustration requires long-horizon coherence beyond single-image text-to-image generation, including narrative decomposition and persistent character identity, layout, and affect across frames. We propose Story-to-Executable Descriptions (S2ED), a training-free, model-agnostic, prompt-layer framework that converts a full story into a sequence of explicit, editable executable descriptions for more consistent rendering. S2ED...

    arxiv.org/abs/2605.22448 · PDF

  29. 29

    Epicure: Navigating the Emergent Geometry of Food Ingredient Embeddings

    Jakub Radzikowski, Josef Chen

    cs.AI · cs.CL · cs.CY

    We present Epicure, a family of three sibling skip-gram ingredient embeddings retrained from scratch on a multilingual recipe corpus. We aggregate 4.14M recipes from 11 sources spanning seven languages, English, Chinese, Russian, Vietnamese, Spanish, Turkish, Indonesian, German, and Indian-English, and normalise the raw ingredient strings to 1,790 canonical entries via an LLM-augmented pipeline. A 203,508-edge ingredient-ingredient NPMI graph...

    arxiv.org/abs/2605.22391 · PDF

  30. 30

    Scaling Observation-aware Planning in Uncertain Domains

    Adrian Zvizdenco, Arthur Conrado Veiga Bosquetti, Alberto Lluch Lafuente, Christoph Matheja

    cs.AI

    Deciding which sensing capabilities to deploy on an agent in uncertain domains is a fundamental engineering challenge, in which one balances task achievability against the high costs of hardware and processing. This problem has previously been formalized as the Optimal Observability Problem (OOP), based on the well-known Partially Observable Markov Decision Process (POMDP) model for decision-making. This work studies (sub-)symbolic techniques...

    arxiv.org/abs/2605.22364 · PDF

  31. 31

    Meta-Soft: Leveraging Composable Meta-Tokens for Context-Preserving KV Cache Compression

    Wei Luo, Yi Huang, Songchen Ma, Huanyu Qu, Jiang Cai, Mingkun Xu

    cs.AI

    The KV cache used in large language models has linearly growing time complexity, so LLMs face memory blow-up and reduced decoding efficiency when they process long contexts.Current KV Cache eviction has become an important research direction; however, existing methods based on fixed Soft Tokens (e.g., Judge Q) rely on a static parameter set as the query to evaluate the importance of KV pairs, so they cannot adapt dynamically to different...

    arxiv.org/abs/2605.22337 · PDF

  32. 32

    Evaluation of Pipelines for Data Integration into Knowledge Graphs

    Marvin Hofer, Erhard Rahm

    cs.AI · cs.DB · cs.LG

    Integrating new data into knowledge graphs (KG) typically involves different tasks that are executed within workflows or pipelines There are many possible pipelines for a specific integration problem but there is not yet a general approach to evaluate the overall quality and performance of such pipelines to be able to determine the best choices. We therefore propose a new benchmark KGI-Bench to evaluate integration pipelines that ingest...

    arxiv.org/abs/2605.22304 · PDF

  33. 33

    Cross-domain benchmarks reveal when coordinated AI agents improve scientific inference from partial evidence

    Fiona Y. Wong, Markus J. Buehler

    cs.AI · cs.LG · cs.MA

    Scientific evidence often spans instruments, databases, and disciplines, so no single source records the full phenomenon. This makes it difficult to determine when coordinated AI agents add value over simpler scientific workflows. We evaluate this question with a cross-domain benchmark spanning four scientific tasks: mapping molecular structure into musical representations, detecting historical paradigm shifts in science, identifying...

    arxiv.org/abs/2605.22300 · PDF

  34. 34

    SciCore-Mol: Augmenting Large Language Models with Pluggable Molecular Cognition Modules

    Yuxuan Chen, Changwei Lv, Yunduo Xiao, Zhongjing Du, Daquan Zhou, Yukun Yan, Zheni Zeng, Zhiyuan Liu

    cs.AI

    Large Language Models (LLMs) are central to the one-for-all intelligent paradigm, but they face a fundamental challenge when dealing with heterogeneous scientific data such as molecules: the inherent gap between discrete linguistic symbols and topological molecular or continuous reaction data leads to significant information loss and semantic noise in text-based reasoning. We propose SciCore-Mol, a modular framework that bridges this gap...

    arxiv.org/abs/2605.22287 · PDF

  35. 35

    Unlocking Proactivity in Task-Oriented Dialogue

    Hongbin Zhang, Ning Gao, Yuqin Dai, Ruiyuan Wu, Jinpeng Wang, Rena Wei Gao, Bingdong Tan, Shuzheng Gao, Zongjie Li,...

    cs.AI

    Proactive task-oriented dialogue (TOD), such as outbound sales, demands a persuasive agent that actively probes the user's concerns and steers the conversation toward acceptance within a bounded number of turns. Yet post-trained LLMs are inherently conservative, and reward-shaping RL (e.g., GRPO) struggles since it only re-weights what an already passive policy samples. We show that conditioning on the user's latent concerns unlocks proactive...

    arxiv.org/abs/2605.22240 · PDF

  36. 36

    Evaluating Large Language Models as Live Strategic Agents: Provider Performance, Hybrid Decomposition, and Operational Gaps in Timed Risk Play

    H. C. Ekne

    cs.AI

    Static benchmarks capture only part of how large language models behave in practice. Real systems place models inside repeated loops with time limits, formatting constraints, and failure modes. We study this setting in a timed multi-phase Risk environment with explicit victory targets and repeated planning and execution cycles. In a replicated 32-game cross-provider championship under frozen rules, gemini-3.1-pro-preview won 20 of 32 games...

    arxiv.org/abs/2605.22238 · PDF

  37. 37

    SGR-Bench: Benchmarking Search Agents on State-Gated Retrieval

    Ningyuan Li, Haiyang Shen, Mugeng Liu, Yudong Han, Zhuofan Shi, Sixiong Xie, Yun Ma

    cs.AI

    Recent advances in large language models and tool-using agents have expanded the range of benchmarked web tasks. Yet an important class of specialized retrieval tasks remains undercharacterized. On many specialized data-retrieval websites, answer-bearing evidence becomes accessible only after establishing the correct site-specific retrieval state through filters, views, hierarchies, or scopes. We term this capability state-gated retrieval...

    arxiv.org/abs/2605.22219 · PDF

  38. 38

    Towards a compositional semantics for quantitative confidence assessment in assurance arguments

    Benjamin Herd, Jessica Kelly, Jan Sabsch, Lydia Gauerhof

    cs.AI

    Assurance arguments provide a clear and structured way to explain why stakeholders should trust that a system satisfies certain properties, yet widely used notations, e.g.Goal Structuring Notation (GSN), typically lack an operational semantics for deriving assurance confidence. Existing approaches address structure and soundness but largely reason over truth values, not over confidence in the justification of claims. Subjective Logic (SL)...

    arxiv.org/abs/2605.22213 · PDF

  39. 39

    CLORE: Content-Level Optimization for Reasoning Efficiency

    Yuyang Wu, Qiyao Xue, Guanxing Lu, Weichen Liu, Zihan Wang, Manling Li, Olexandr Isayev

    cs.AI

    Reinforcement learning post-training has improved the reasoning ability of large language models, but often produces unnecessarily long, repetitive, or semantically opaque reasoning traces. Existing efficient reasoning methods mainly regulate response length through explicit budgets or length-aware rewards, leaving intermediate reasoning content weakly supervised. We propose CLORE, a content-level optimization framework that improves...

    arxiv.org/abs/2605.22211 · PDF

  40. 40

    Skill Weaving: Efficient LLM Improvement via Modular Skillpacks

    Zhuo Li, Guodong Du, Zesheng Shi, Weiyang Guo, Weijun Yao, Yuan Zhou, Jiabo Zhang, Jing Li

    cs.AI · cs.LG

    Large language models increasingly require specialization across diverse domains, yet existing approaches struggle to balance multi-domain capacities with strict memory and inference constraints. In this work, we introduce SkillWeave, a modular improvement framework that enables LLMs to specialize under fixed memory budgets. SkillWeave partitions full capabilities of a general-purpose model into skillpacks -- lightweight, domain-specific...

    arxiv.org/abs/2605.22205 · PDF

  41. 41

    LLM-Metrics: Measuring Research Impact Through Large Language Model Memory

    Si Shen, Wenhua Zhao, Danhao Zhu

    cs.AI

    Citation counts remain the dominant metric for assessing research impact, yet they suffer from well-documented limitations: temporal lag, disciplinary bias, and Matthew effects. Here we propose LLM-Metrics, a research-impact assessment metric derived from the parametric memory of large language models (LLMs). The central hypothesis is that high-impact papers receive greater exposure in the academic community, that this exposure enters LLM...

    arxiv.org/abs/2605.22176 · PDF

  42. 42

    Measuring Cross-Modal Synergy: A Benchmark for VLM Explainability

    Joël Roman Ky, Salah Ghamizi, Maxime Cordy

    cs.AI · cs.LG

    Vision-Language Models (VLMs) map complex visual inputs to semantic spaces, but interpreting the cross-modal reasoning of VLMs currently relies on post-hoc explainers evaluated via unimodal perturbation metrics. We expose a limitation in this paradigm: because multimodal datasets contain language priors and modality biases, VLMs frequently exhibit cross-modal redundancy, allowing them to answer visual queries using text alone. Consequently,...

    arxiv.org/abs/2605.22168 · PDF

  43. 43

    Adapting the Interface, Not the Model: Runtime Harness Adaptation for Deterministic LLM Agents

    Tianshi Xu, Huifeng Wen, Meng Li

    cs.AI

    LLM agents are shaped not only by their language models, but also by the runtime harness that mediates observation, tool use, action execution, feedback interpretation, and trajectory control. While existing agent adaptation methods mainly update model parameters, many failures in deterministic, rule-governed domains stem from mismatches at the model--environment interface. We propose Life-Harness, a lifecycle-aware runtime harness that...

    arxiv.org/abs/2605.22166 · PDF

  44. 44

    ST-SimDiff: Balancing Spatiotemporal Similarity and Difference for Efficient Video Understanding with MLLMs

    Bingjun Luo, Tony Wang, Chaoqi Chen, Xinpeng Ding

    cs.AI · cs.CV

    Multimodal Large Language Models (MLLMs) face significant computational overhead when processing long videos due to the massive number of visual tokens required. To improve efficiency, existing methods primarily reduce redundancy by pruning or merging tokens based on importance or similarity. However, these approaches largely overlook a critical dimension of video content, i.e., changes and turning points, and they lack a collaborative model...

    arxiv.org/abs/2605.22158 · PDF

  45. 45

    IdleSpec: Exploiting Idle Time via Speculative Planning for LLM Agents

    Daewon Choi, Kyunghyun Park, Woomin Song, Saket Dingliwal, Sai Muralidhar Jayanthi, Jinwoo Shin, Aram Galstyan

    cs.AI

    Large language model (LLM)-based agents solve complex tasks by leveraging multi-step reasoning with iterative tool calls and environment interactions, which incur idle time while waiting for observations. Despite the prevalence of idle time in most agentic scenarios, existing works treat it as an unavoidable overhead or propose restricted solutions that overlook varying computational budgets across different tool calls and future observation...

    arxiv.org/abs/2605.22154 · PDF

  46. 46

    Ratchet: A Minimal Hygiene Recipe for Self-Evolving LLM Agents

    Xing Zhang, Yanwei Cui, Guanghui Wang, Ziyuan Li, Wei Qiu, Bing Zhu, Peiyang He

    cs.AI · cs.CL

    Self-evolving skill libraries, pioneered by Voyager, let frozen LLM agents accumulate reusable knowledge without weight updates, yet recent evaluation shows that LLM-authored skills deliver $+0.0$pp over no-skill baselines while human-curated ones deliver $+16.2$pp: the bottleneck is not skill authoring but lifecycle management. We introduce \textbf{Ratchet}, a single-agent loop in which a frozen LLM writes, retrieves, curates, and retires...

    arxiv.org/abs/2605.22148 · PDF

  47. 47

    Efficient Agentic Reasoning Through Self-Regulated Simulative Planning

    Mingkai Deng, Jinyu Hou, Lara Sá Neves, Varad Pimpalkhute, Taylor W. Killian, Zhengzhong Liu, Eric P. Xing

    cs.AI · cs.CL · cs.LG · cs.RO

    How should an agent decide when and how to plan? A dominant approach builds agents as reactive policies with adaptive computation (e.g., chain-of-thought), trained end-to-end expecting planning to emerge implicitly. Without control over the presence, structure, or horizon of planning, these systems dramatically increase reasoning length, yielding inefficient token use without reliable accuracy gains. We argue efficient agentic reasoning...

    arxiv.org/abs/2605.22138 · PDF

  48. 48

    Perception or Prejudice: Can MLLMs Go Beyond First Impressions of Personality?

    Caixin Kang, Tianyu Yan, Sitong Gong, Mingfang Zhang, Liangyang Ouyang, Ruicong Liu, Bo Zheng, Huchuan Lu, Kaipeng...

    cs.AI · cs.CV · cs.CY

    Multimodal Large Language Models (MLLMs) are increasingly deployed in human-facing roles where personality perception is critical, yet existing benchmarks evaluate this capability solely on numerical Big Five score prediction, leaving open whether models truly perceive personality through behavioral understanding or merely prejudge through superficial pattern matching. We address this gap with three contributions. (i) A new task: we formalize...

    arxiv.org/abs/2605.22109 · PDF

  49. 49

    ArborKV: Structure-Aware KV Cache Management for Scaling Tree-based LLM Reasoning

    Yeqiu Chen, Ziyan Liu, Zhenxin Huang, Runquan Gui, Hong Wang, Lei Liu

    cs.AI

    Recent progress in LLM reasoning has increasingly shifted from single-pass generation to explicit search over intermediate reasoning states. Tree-of-Thoughts (ToT) organizes inference to tree-structured search with branching and backtracking, but it substantially amplifies the Key--Value (KV) cache: retaining KV states for a frontier of partial trajectories quickly becomes a memory bottleneck that limits throughput and constrains search depth...

    arxiv.org/abs/2605.22106 · PDF

  50. 50

    ExComm: Exploration-Stage Communication for Error-Resilient Agentic Test-Time Scaling

    Woomin Song, Beomjun Kim, Daewon Choi, Sai Muralidhar Jayanthi, Saket Dingliwal, Jinwoo Shin, Aram Galstyan

    cs.AI

    A common failure mode in long-horizon agentic test-time scaling is error propagation, where factual errors or invalid deductions introduced at intermediate steps persist in the agent's belief state and contaminate later reasoning. Existing test-time scaling methods provide limited control over this process, as they often rely on agents to detect their own mistakes, select among flawed trajectories, or refine solutions only after errors have...

    arxiv.org/abs/2605.22102 · PDF

  51. 51

    MPDocBench-Parse: Benchmarking Practical Multi-page Document Parsing

    Bangbang Zhou, Hangdi Xing, Yifan Chen, Jianjun Xu, Qi Zheng, Feiyu Gao, Zhibo Yang, Shuai Bai, Ming Yan, Jieping...

    cs.AI

    Document parsing converts visually rich documents into machine-readable structured representations, forming a crucial foundation for information systems. Although many benchmarks have been proposed for document parsing, they remain inadequate for realistic scenarios. Existing benchmarks either focus on specific tasks or assess only single-page, text-centric settings, making them insufficient for practical multi-page parsing. Moreover, they...

    arxiv.org/abs/2605.22100 · PDF

  52. 52

    Knowledge Graph Re-engineering Along the Ontological Continuum (extended version)

    Enrico Daga, Valentina Tamma, Terry Payne

    cs.AI

    Knowledge graphs have become the primary vehicle for data integration and are critical to the success of modern AI, but the diversity of KG modelling practices, from lightweight vocabularies to richly axiomatised ontologies, makes integration and reuse expensive and brittle. This challenge is particularly acute in neuro-symbolic AI, where bridging neural and symbolic components depends on the ability to reengineer KGs to fit new requirements;...

    arxiv.org/abs/2605.22093 · PDF

  53. 53

    A Camera-Cooperative ISAC Framework for Multimodal Non-Cooperative UAVs Sensing

    Wenfeng Wu, Luping Xiang, Kun Yang

    cs.AI

    The detection of non-cooperative unmanned aerial vehicles (UAVs) presents significant challenges for Integrated Sensing and Communication (ISAC) systems due to the inherent limitations of single-modal perception and the competition for shared communication and sensing resources. To address these challenges, this paper proposes a novel Camera-Cooperative ISAC (CC-ISAC) framework that employs multimodal sensing to enable efficient UAV beam...

    arxiv.org/abs/2605.22090 · PDF

  54. 54

    Enhancing Visual Token Representations for Video Large Language Models via Training-Free Spatial-Temporal Pooling and Gridding

    Bingjun Luo, Tony Wang, Hanqi Chen, Xinpeng Ding

    cs.AI · cs.CV

    Recent advances in Multimodal Large Language Models (MLLMs) have significantly advanced video understanding tasks, yet challenges remain in efficiently compressing visual tokens while preserving spatiotemporal interactions. Existing methods, such as LLaVA family, utilize simplistic pooling or interpolation techniques that overlook the intricate dynamics of visual tokens. To bridge this gap, we propose ST-GridPool, a novel training-free visual...

    arxiv.org/abs/2605.22078 · PDF

  55. 55

    Active Evidence-Seeking and Diagnostic Reasoning in Large Language Models for Clinical Decision Support

    Chen Zhan, Xihe Qiu, Xiaoyu Tan, Xibing Zhuang, Gengchen Ma, Yue Zhang, Shuo Li, Peifeng Liu, Xiaoxiao Ge, Liang Liu, Lu Gan

    cs.AI

    Large language models perform well on static medical examinations, yet clinical diagnosis often requires iterative evidence gathering under uncertainty. Building on prior interactive evaluation efforts, we introduce an OSCE-inspired standardized patient simulator and a controlled, reproducible benchmark for active diagnostic inquiry. Across 468 cases and 15 models in our protocol, we observe that multi-turn evidence seeking reduces diagnostic...

    arxiv.org/abs/2605.22047 · PDF

  56. 56

    The Log is the Agent: Event-Sourced Reactive Graphs for Auditable, Forkable Agentic Systems

    Yohei Nakajima

    cs.AI · cs.MA

    Most agent frameworks are built around the language model: a conversation loop comes first, then tools, then rules, and finally a logging layer bolted on for observability, with state persisted as retrievable "memory." We describe ActiveGraph, a runtime that inverts this arrangement. The append-only event log is the source of truth; the working graph is a deterministic projection of that log; and behaviors--ordinary functions, classes,...

    arxiv.org/abs/2605.21997 · PDF

  57. 57

    ECPO: Evidence-Coupled Policy Optimization for Evidence-Certified Candidate Ranking

    Miaobo Hu, Shuhao Hu, BoKun Wang, Yina Sa, Xin Wang, Xiaobo Guo, Daren Zha, Jun Xiao

    cs.AI · cs.LG

    Ranking systems used in decision-support settings should not only order candidates but also expose evidence that can be independently checked. We study evidence-certified candidate ranking: given an intent_id, a predefined plan skeleton, a window-local candidate roster, and text-derived candidate trajectories with span provenance, a system must output a Top-K list together with doc_id:span evidence certificates whose cited spans are...

    arxiv.org/abs/2605.21993 · PDF

  58. 58

    Echo: Learning from Experience Data via User-Driven Refinement

    Hande Dong, Xiaoyun Liang, Jiarui Yu, Jiayi Lin, Changqing Ai, Feng Liu, Wenjun Zhang, Rongbi Wei, Chaofan Zhu,...

    cs.AI · cs.CL

    Static "human data" faces inherent limitations: it is expensive to scale and bounded by the knowledge of its creators. Continuous learning from "experience data" - interactions between agents and their environments - promises to transcend these barriers. Today, the widespread deployment of AI agents grants us low-cost access to massive streams of such real-world experience. However, raw interaction logs are inherently noisy, filled with...

    arxiv.org/abs/2605.21984 · PDF

  59. 59

    Format-Constraint Coupling in Knowledge Graph Construction from Statistical Tables

    Jingxuan Qi, Zhiqiang Ye, Yuxiang Feng

    cs.AI

    An extraction schema should not reduce knowledge graph fidelity. On statistical CSV, however, it can. We study country-by-year time-series matrices, a common layout on open-data portals. In this setting, serialization format and schema constraints interact super-additively. Their joint effect exceeds the sum of independent effects by up to +1.180 (2x2 factorial, 6 datasets). Bootstrap 95% CIs are strictly positive on 4/6 datasets, with...

    arxiv.org/abs/2605.21974 · PDF

  60. 60

    AI-Enabled Serious Games: Integrating Intelligence and Adaptivity in Training Systems

    Priyamvada Tripathi, Bill Kapralos

    cs.AI · cs.CY · cs.HC · cs.MA

    Serious games are widely used for learning and training across domains such as healthcare, defense, and education. Persistent challenges remain, however, including static scenario design, authoring bottlenecks, limited learner modeling, and difficulty implementing meaningful real-time instructional adaptation. Recent advances in artificial intelligence (AI) introduce novel capabilities such as dynamic scenario variation, contextual feedback,...

    arxiv.org/abs/2605.21962 · PDF

  61. 61

    Planning in the LLM Era: Building for Reliability and Efficiency

    Michael Katz, Harsha Kokel, Kavitha Srinivas, Shirin Sohrabi

    cs.AI · cs.CL

    Growing attention to intelligent agents has put a spotlight on one of their central capabilities: planning. Early attempts to leverage large language models (LLMs) for planning relied on single-shot plan generation, followed by hybrid approaches that coupled LLMs with limited external search. These methods, unsound and incomplete by their very nature, often require substantial resources without yielding better solutions on unseen problems. As...

    arxiv.org/abs/2605.21902 · PDF

  62. 62

    FLUID: From Ephemeral IDs to Multimodal Semantic Codes for Industrial-Scale Livestreaming Recommendation

    Xinhang Yuan, Zexi Huang, Anjia Cao, Xudong Lu, Zikai Wang, Penghao Zhou, Chang Liu, Wentao Guo, Qinglei Wang

    cs.AI

    Modern recommender systems rely heavily on ID-based collaborative filtering: each item is represented by a unique ID embedding that accumulates collaborative signals from user interactions. Livestreaming recommendation, however, faces a unique challenge in this paradigm: a live room typically broadcasts for only tens of minutes, so its item ID remains poorly learned in a persistent cold-start state and ID-centric ranking models fail to...

    arxiv.org/abs/2605.21832 · PDF

  63. 63

    Toward AI VIS Co-Scientists: A General and End-to-End Agent Harness for Solving Complex Data Visualization Tasks

    Haichao Miao, Zhimin Li, Kuangshi Ai, Kaiyuan Tang, Chaoli Wang, Peer-Timo Bremer, Shusen Liu

    cs.AI · cs.HC

    The ability to inspect, interpret, and communicate complex data is crucial for virtually any scientific endeavor, but often requires significant expertise outside the core domain ranging from data management and analysis to visualization design and implementation. We present an end-to-end agentic harness that, based on only the data and a high level description of the tasks, independently designs custom visual analysis applications (VIS...

    arxiv.org/abs/2605.21825 · PDF

  64. 64

    Implicit Safety Alignment from Crowd Preferences

    Qian Lin, Daniel S. Brown

    cs.AI

    Reinforcement Learning from Human Feedback (RLHF) can reveal implicit objectives such as safety considerations that go beyond task completion. In this work, we focus on the common safety criteria embedded in crowd preference datasets, where different users may express distinct preferences or objectives, yet follow similar safety principles. Our aim is to discover shared safety criteria from crowd preferences and then transfer them to...

    arxiv.org/abs/2605.21822 · PDF

  65. 65

    Trace2Skill: Verifier-Guided Skill Evolution for Long-Context EDA Agents

    Zijian Du, Nathaniel Pinckney

    cs.AI · cs.MA

    Complex Verilog Design Problems (CVDP) challenge hardware LLM agents because solving them requires localizing verifier-relevant RTL, testbenches, include paths, and build dependencies inside large repository snapshots, making precise edits, and recovering from sparse hidden-verifier failures. We present Trace2Skill, a test-time scaling framework that improves a hardware agent without RTL-specialized model fine-tuning. Rather than training a...

    arxiv.org/abs/2605.21810 · PDF

  66. 66

    What Counts as AI Sycophancy? A Taxonomy and Expert Survey of a Fragmented Construct

    Meryl Ye, Lujain Ibrahim, Jessica Y. Bo, Myra Cheng, Ida Mattsson, Daniel Vennemeyer, Robert Kraut, Steve Rathje

    cs.AI

    AI sycophancy has become a prominent concern in large language model (LLM) research. Yet the term lacks a consistent definition and has been applied to behaviors ranging from agreeing with a user's false claim to excessively praising the user to withholding corrective feedback. When researchers, companies, and policymakers use the same term to describe different behaviors, evaluation results become difficult to compare, mitigation strategies...

    arxiv.org/abs/2605.21778 · PDF

  67. 67

    A Causal Argumentation Method for Explainability of Machine Learning Models

    Henry Salgado, Meagan R. Kendall, Martine Ceberio

    cs.AI

    Explainable AI (XAI) methods identify which features are relevant to a model's predictions but often fail to clarify why certain decisions are made. In this work, we present a novel method that integrates causality with argument-based reasoning to explain why models may be making predictions. Our approach first identifies causal relationships among variables using causal discovery methods and then translates these into a Bipolar Argumentation...

    arxiv.org/abs/2605.21758 · PDF

  68. 68

    Who Uses AI? Platforms, Workforce, and AI Exposure

    Michelle Yin, Burhan Ogut

    cs.AI · econ.GN

    A growing literature uses artificial intelligence platform conversation logs to measure occupation exposure. We show that these scores partly measure platform user base rather than the workforce. Holding outcome, sample, controls, and estimator fixed while varying only the platform input changes the post-ChatGPT employment coefficient by a factor of 1.9, and within-vendor consumer-versus-enterprise channels produce estimates that disagree in...

    arxiv.org/abs/2605.21743 · PDF

This edition is part of The Daily Abstract — cs.AI archive. Subscribe to receive these in your inbox each morning, automatically translated to Spanish, with reply-to-PDF: arxivdaily.ignorelist.com.

Colophon Set in Georgia, with system sans for interface chrome and a monospaced stack for code and paper identifiers. Sole accent: amber #D99C5E. Built and served on an always-free VM. The masthead is set 14% letterspaced because newspapers do that and it works.