cs.AI · 2026-10-19 · No. 148

Artificial Intelligence, 2026-10-19.

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

01 — The papers

51 entries
  1. 01

    AI-Driven Reproducibility and Analytical Robustness in Scientific Discovery

    Shicheng Liu, Krista Opsahl-Ong, Andrew Drozdov, Julian Killingback, Arnav Singhvi, Ronak Pradeep, Michael...

    cs.AI · cs.CL

    Reproducible and analytically robust research findings are essential to scientific progress, but evaluating these properties typically requires costly manual effort. We investigate whether LLM agents can help automate reproduction and robustness analysis, using a medical case study grounded in the Surveillance, Epidemiology, and End Results (SEER) database, a widely used U.S. cancer registry underlying tens of thousands of published studies....

    arxiv.org/abs/2610.23845 · PDF

  2. 02

    Self-Building Labs

    Jiaru Bai, Yuchi Zhao, Han Hao, Miroslav Bogdanovic, Yang Cao, Varinia Bernales, Alán Aspuru-Guzik

    cs.AI · cs.MA · cs.RO

    Self-driving laboratories (SDLs) accelerate scientific discovery by automating experimental execution and optimization. AI agents can increasingly propose new experiments, but existing SDLs can only run the procedures they were built for. Adapting them to a new experiment still takes extensive manual engineering, which makes laboratory flexibility the bottleneck for scaling discovery. We introduce the self-building laboratory (SBL), which...

    arxiv.org/abs/2610.23826 · PDF

  3. 03

    Beyond SAEs: Consistent Concept Discovery with Sparse Coding

    Jinho Choi, Hyesu Lim, Jaegul Choo, Steffen Schneider

    cs.AI

    Sparse autoencoders (SAEs) explain foundation model representations by decomposing activations into sparse combinations of learned dictionary atoms. However, the learned dictionaries can vary across training runs, limiting the reliability and reproducibility of SAE-based analyses. We propose Consistent Sparse Dictionary Learning (ConSDL), which replaces the standard linear encoder with per-sample sparse coding. This is motivated by our...

    arxiv.org/abs/2610.23801 · PDF

  4. 04

    FuTransformer: FP8 prefill and INT4 decode for Llama-style inference on Hopper

    Wei Wang

    cs.AI

    Transformer prefill reuses weights across tokens and is dominated by matrix multiplication and activation data movement; single-token decode has limited reuse and is dominated by weight and KV-cache reads. We present FUTRANSFORMER, a CUDA implementation of a Llama-style Transformer block for NVIDIA Hopper GPUs. It applies CODA's GEMM-plus-epilogue approach to inference, adding direct KV-cache placement, folding the fixed RMSNorm gain into...

    arxiv.org/abs/2610.23796 · PDF

  5. 05

    DrugHarness: An Evolvable Agent Harness for Computational Drug Discovery

    Yuxing Chen, Zhisheng Zhang, Cheng Tang, Wei Li, Minghui Zhang, Ziyan Huang, Shaocong Wei, Jiawei Chen, Junru Hu,...

    cs.AI

    Computational drug discovery requires agents to compose scientific operations and adapt to changing evidence and constraints. Here we introduce DrugHarness, an evolvable agent harness coupling composable scientific execution with conditional skill induction. Its environment integrates over 80 scientific operations, biomedical resources and task-specific code, preserving associations among candidates, results and artifacts. From execution...

    arxiv.org/abs/2610.23791 · PDF

  6. 06

    Beyond Large Language Models: A Mental World Model Framework for Mental Modeling, Simulation, Evolution, and Interpretability

    Qingshi Wang

    cs.AI

    Understanding and simulating how subjective cognition and mental states emerge, evolve, and shape behavior is essential for advancing artificial intelligence toward deeper social cognition and general intelligence. Existing large language models (LLMs) and Theory of Mind (ToM) methods primarily infer mental states from linguistic information, lacking unified mechanisms for subjective world construction, cognitive-psychological interactions,...

    arxiv.org/abs/2610.23722 · PDF

  7. 07

    OwlSight: A Panoramic Monitor for Chain-of-Thought Reasoning

    Lewen Yan, Yanxu Zhu, Jie Zhang, Jing Shao

    cs.AI

    Chain-of-Thought (CoT) monitoring offers actionable evidence for oversight AI systems, with the potential to enable early detection and intervention. However, existing approaches are focused on specific tasks, whereas the regulation of AI, particularly reasoning models, requires monitoring a broader range of tasks. We introduce Factor for CoT monitoring, a strategy that supports customizing monitoring requirements in the prompt, which enables...

    arxiv.org/abs/2610.23670 · PDF

  8. 08

    GEMS: Molecular Structure Elucidation as Learned Search over Isomer Graphs

    Utku Umur Acikalin, Yingheng Wang, Goncalo J. Gouveia, Tyler J. Schwertfeger, Frank C. Schroeder, Carla P. Gomes

    cs.AI

    Identifying small-molecule structures from molecular formulas and tandem mass spectra remains a bottleneck in untargeted metabolomics and natural-product discovery. Even when database retrieval or de novo prediction misses the exact structure, the resulting candidates can provide starting points for exploring nearby structures through a few bond edits. We introduce GEMS (Guided Editing of Molecular Structures), which refines candidates...

    arxiv.org/abs/2610.23617 · PDF

  9. 09

    From Trajectories to Differentiated Skills: Cognitive Differentiation for Self-Improving Agents

    Yewei Zhao, Derong Xu, Wei Chen, Yin Xu, Peilun Zhou, Xiaohan Wang, Guojun Yin, Wei Lin, Mingjun Xiao

    cs.AI

    Large language model (LLM) agents tackle complex tasks through reasoning and interaction with external environments. Agent skills provide persistent, transferable capabilities that guide task planning, inform contextual decisions, support tool use, and enable executable procedures. Yet existing skill-evolution methods largely organize knowledge around skill artifacts, overlooking the functional heterogeneity within individual skills and its...

    arxiv.org/abs/2610.23599 · PDF

  10. 10

    RL-Guided Temporal Localization for Dual-Channel Retrieval in Long-Horizon Agent Memory

    Dabin Sheng, Zhe Wu, Jinming Zhao, Junliang Xing, Yuanchun Shi

    cs.AI

    Time-aware retrieval is crucial for long-horizon LLM agents to effectively leverage ever-growing long-term memory. Existing methods focus on lexical or semantic similarity, leading to target memories being crowded out by temporally mismatched yet semantically similar candidates. Even time-aware approaches inject temporal cues only during memory organization or answer generation, overlooking temporal constraints at the retrieval stage, where...

    arxiv.org/abs/2610.23598 · PDF

  11. 11

    Darwin-Claw: Environment-Adaptive Experience Consolidation for Recursively Self-Evolving Agents

    Mingze Yin, Xiaohan Wang, Dian Li, Longfei Luo, Jiahong Yan, Yiheng Zhu, Chang-Yu Hsieh

    cs.AI

    While Large Language Model (LLM) agents have demonstrated remarkable capabilities across complex tasks, they predominantly operate in isolated sessions, lacking the ability to continually learn from historical interactions or adapt to user-specific environments. Existing self-evolution paradigms often indiscriminately inject raw interaction trajectories into memory or skill modules, leading to obscured behavioral patterns and compromised...

    arxiv.org/abs/2610.23574 · PDF

  12. 12

    SkillBrowser: Selective Skill Exposure for Scalable Skill-Using Agents

    Jin Ye, Yifan Jiang, Lan-zhe Guo

    cs.AI

    Large language model agents increasingly rely on skills to extend their capabilities in complex tasks. However, scalable skill use requires Selective Skill Exposure (SSE): providing relevant procedural content when needed while avoiding unnecessary exposure that may disrupt reasoning. Existing paradigms fall short in complementary ways. Progressive disclosure exposes metadata for a growing skill library, introducing context pollution;...

    arxiv.org/abs/2610.23554 · PDF

  13. 13

    Trapped in Your Story: Evaluating Narrative-Induced Sycophancy in Large Language Models

    Runtai Yang, Dandan Zhu, Kaiwei Zhang, Guangtao Zhai

    cs.AI

    Large language models (LLMs) are known to exhibit sycophancy by adjusting their responses toward users' expressed views. However, real users often justify their opinions through reasons, personal experiences, causal explanations, and broader narratives, raising the question of whether such narratives can further steer model judgments. We construct a controlled benchmark of 500 debatable questions across ten domains, where the user stance is...

    arxiv.org/abs/2610.23525 · PDF

  14. 14

    LocalIntent: Outcome-Specific Structured Generation for Backend-Compatible Smart Building Automation with Local LLMs

    Tianyang Liu, Jiayue Wang, Zhanhao Cheng, Changyang Wang, Shaoshu Zhu, Zhenghao Wu, Hadi Tabatabaee Malazi

    cs.AI

    Locally deployed large language models (LLMs) can reduce reliance on remote cloud services for natural language smart building automation, but backend-facing processing requires correct outcome selection, alignment with local capabilities, and valid structured outputs. We investigate whether a complete outcome-specific prompting design improves compact-model outputs beyond generic route-then-generate prompting. The proposed LocalIntent...

    arxiv.org/abs/2610.23511 · PDF

  15. 15

    S-GAP: Stealthy Graph Attack Primitives for Graph Large Language Models

    Wenda Li, Tongya Zheng, Shunyu Liu, Hanyang Yuan, Kaixuan Chen, Yu Wang, Yuchen Ying, Haoze Jiang, Huan Li, Gang...

    cs.AI

    Graph Large Language Models (GraphLLMs) enable powerful semantic understanding and reasoning over Text-Attributed Graphs (TAGs), while their semantic reasoning may exhibit vulnerabilities to perturbations in node text and graph structure under adversarial attack evaluation. Existing attacks typically construct textual and structural perturbations in separate stages guided by distinct realization intents. However, this separation results in...

    arxiv.org/abs/2610.23465 · PDF

  16. 16

    GatewayBench: Unified Capability and Security Evaluation Platform for MCP Gateways

    Yitian Zheng, Zhaorun Chen, Haoyu Li, Bo Li

    cs.AI

    AI agents have increasingly interacted with external environments by invoking tools, with Model Context Protocol (MCP) emerging as a standard interface for connecting agents to heterogeneous tools and services. As organizations deploy agents at scale, MCP gateways have become a critical middleware layer for aggregating, governing, and shaping the tool surface exposed to agents. Despite the growing importance of MCP gateways in agentic...

    arxiv.org/abs/2610.23449 · PDF

  17. 17

    ScoutRoute: Efficient Mixture-of-Experts Routing Using Internal Language Model Signals

    Rajdeep Pandhare, Delzaan Bharucha

    cs.AI · cs.CL

    Efficient expert routing is a fundamental challenge in Mixture-of-Experts (MoE) systems, particularly for heterogeneous collections of independently trained Large Language Mod els (LLMs). Existing routing approaches typically rely on jointly trained gating networks, prompt-based predictors, or computationally expensive multi-model inference, limiting their flexibility and efficiency. This paper presents ScoutRoute, a lightweight post-hoc...

    arxiv.org/abs/2610.23443 · PDF

  18. 18

    Generative Policy Improvement from Observational Data

    Jonas Schweisthal, Dennis Frauen, Stefan Feuerriegel, Emma Brunskill

    cs.AI · cs.LG

    We study how to improve decision policies from observational data to generate better actions when actions are complex, structured objects, such as text messages or images. This task is subject to two major challenges: (i) action-level overlap is often too weak for reliable off-policy evaluation and improvement, and (ii) generating new structured actions often requires expensive training or fine-tuning (e.g., of an LLM) and can be ill-defined...

    arxiv.org/abs/2610.23436 · PDF

  19. 19

    Monitor-Validity Failures: LLM Safety Evidence Paths at Evolving Framework Boundaries

    Junchi Liao

    cs.AI · cs.CR · cs.SE

    Modern LLM applications rely on runtime monitors that inspect tool calls, traces, streams, detector outputs, and verifier inputs. This creates an interface-reliability problem: a framework path may remain operational while policy-relevant evidence never reaches its decision point. In one representative case, a regex detector returned a positive match, but the framework stranded the result before the block decision. This differs from...

    arxiv.org/abs/2610.23362 · PDF

  20. 20

    OliO: ODE-based Linear Transition Operator for Self-Supervised Time Series Forecasting

    Kwangryeol Park, Seulki Lee

    cs.AI

    Self-supervised representation learning has advanced time series forecasting by capturing robust features from unlabeled data. However, existing contrastive and masking-based methods often struggle to explicitly model the underlying temporal flow, either by treating time lag as noise or by disrupting the inherent temporal dynamics essential for forecasting. To address these limitations, we propose OliO (**O**DE-based **Li**near Transition...

    arxiv.org/abs/2610.23345 · PDF

  21. 21

    Robust Function Calling for LLM Agents Under Interface Drift

    Aniruddha Chattopadhyay, Kanishka Halder, Partha Sarathi Roy

    cs.AI · cs.CL · cs.SE

    Tool-augmented language-model agents rely on interface text: tool descriptions, argument schemas, and the candidate tool list shown at inference time. These interfaces often change after deployment even when the underlying tool semantics remain fixed, creating a practical reliability problem for function-calling systems. We study this failure mode as interface drift and evaluate whether a lightweight, training-free repair layer can recover...

    arxiv.org/abs/2610.23342 · PDF

  22. 22

    AgentOnto: Ontology-Guided LLM Agents for Enterprise Data Analytics

    Dayan Pan, Bincheng Zuo, Xiaojian Wang, Daogang Zhang, Jingyuan Wang

    cs.AI · cs.DB

    Enterprise data analysis requires agents to connect analytical requests with knowledge distributed across databases, business documents, and existing example tasks. Without an explicit representation of this knowledge, an agent may produce incorrect analytical results and cannot reliably incorporate feedback. We present AgentOnto, an ontology-guided LLM agent for enterprise data analytics. AgentOnto organizes heterogeneous enterprise...

    arxiv.org/abs/2610.23317 · PDF

  23. 23

    From Memory Hits to Answerable Evidence: Query-Conditioned Memory Compilation for Long-Term Agent Memory

    Shangze Li, Yanyu Yao, Derong Xu, Zhi Zheng, Tong Xu

    cs.AI

    Long-term agent memory requires recovering answer-relevant evidence dispersed across turns and sessions. Existing systems increasingly use structured memories and adaptive retrieval, yet often pass retrieved or expanded units directly to the reader, coupling memory addressing with evidence presentation and leaving the final context redundant or incomplete. To address this gap, we propose MemCompile, a query-conditioned evidence compilation...

    arxiv.org/abs/2610.23313 · PDF

  24. 24

    Self-Reflective Chain-of-Thought (SRCoT): Evidence-Based Text Generation via a Closed-Loop Retrieve--Reflect--Revise Pipeline on Edge LLMs

    Fabian Broszio, Michael Hoffmann, Adrian Paschke

    cs.AI · cs.CL

    Large language models (LLMs) can generate fluent outputs that lack source support or conflict with local operational constraints. We introduce Self-Reflective Chain-of-Thought (SRCoT), a modular architecture that organizes generation as a closed Retrieve--Reflect--Revise loop. SRCoT separates environment state, persistent knowledge, accumulated experience, and short-term memory into provenance-labeled stores. BM25 retrieval supplies local...

    arxiv.org/abs/2610.23311 · PDF

  25. 25

    SimOPD: Relational Knowledge Transfer for On-Policy Distillation

    Kai Tang, Xing Zheng, Haobo Wang, Guangcheng Zhu, Shenzhi Yang, Zhengqing Zang, Yingfan Ma, Bowen Song, Weiqiang Wang

    cs.AI · cs.CL

    A key challenge in on-policy distillation (OPD) is extracting effective supervision from teacher knowledge. Supervision from token probabilities may underutilize internal knowledge, while direct hidden-state matching forces the student to shift to the teacher's feature space and complicates cross-architecture transfer. Inspired by contrastive learning's use of relationships between instances, we propose \textbf{SimOPD} to transfer teacher...

    arxiv.org/abs/2610.23306 · PDF

  26. 26

    Optimist Deep Research: An Iterative, State-Centric Agent for Autonomous Literature Research

    Gihan Panapitiya

    cs.AI

    Most retrieval-augmented systems treat literature search as a single retrieval step followed by summarization. Real literature research works differently: the investigator's understanding of the question changes as they read, new subquestions appear, and evidence is weighed, not just collected. We present Optimist Deep Research, an open-source agent that runs literature research as an iterative evidence-acquisition process over an explicit,...

    arxiv.org/abs/2610.23299 · PDF

  27. 27

    CoTA++: Understanding the Repeat Curse in dMLLMs from an Information Flow Perspective

    Qiyan Zhao, Xiaofeng Zhang, Xu-Yao Zhang, Shijie Han, Zhepeng Li, Shunzhi Zhu, Da-Han Wang

    cs.AI

    Diffusion-based multimodal large language models (dMLLMs) decode by iterative denoising with bidirectional attention, which precludes the standard autoregressive key--value cache. Practical deployments therefore rely on approximate caching that reuses intermediate states across denoising steps. Such caching, however, often makes a response repeat one token at adjacent positions in a run that breaks the sentence, a failure we term the...

    arxiv.org/abs/2610.23290 · PDF

  28. 28

    Annotrace: A Workbench for Step-Level Annotation of Agent Traces

    Arun Josephraj Arokiaraj, Chenyi Huang, Sahan Bulathwela, Zekun Wu, Bin Wu

    cs.AI · cs.CL · cs.HC · cs.MA

    Every study of why LLM agents fail begins by labelling traces by hand, and every such study builds its own throwaway tool. Observability platforms record agent traces but offer no structured labelling; annotation tools label text but have no notion of a step, a delegation or a tool call, leaving it unclear what the smallest unit of execution to label should be. Annotrace reads six trace formats, including OpenTelemetry GenAI, LangSmith and...

    arxiv.org/abs/2610.23284 · PDF

  29. 29

    Reasoning Models Can Control Their Chains of Thought When They Tell Themselves To

    Uriel Dolev, Idan Schwartz, Yoav Goldberg

    cs.AI · cs.CL · cs.LG

    Chain-of-thought (CoT) controllability, the ability of reasoning models to control properties of their CoT, is consequential for both their usefulness and reliable oversight. Recent work evaluates this capability by measuring whether models follow CoT constraints given in the user prompt, and reports very low controllability across models. We argue that this low controllability may not reflect an inherent inability of reasoning models to...

    arxiv.org/abs/2610.23257 · PDF

  30. 30

    Beyond Reward Scoring: Online Multi-Agent Preference Learning for Multimodal Response Generation

    Ziyun Zou, Jiwoo Hong, Yifan Lu, Shao Tang, Weiyi Wu, Aman Gupta, Jin Tian, Imran Razzak

    cs.AI

    Multimodal preference alignment often follows a score-based RLHF pipeline: a reward model is first trained to learn human preferences, then used to score on-policy responses and optimize the response model's generation. This turns preference supervision into a scalar signal for generation: a model may learn to produce answers with higher preference scores, but is rarely trained explicitly to learn the reasons underlying human preferences. We...

    arxiv.org/abs/2610.23256 · PDF

  31. 31

    Characterising and Analysing Retrieval-Augmented Generation Techniques

    Emmanuel Igbekele, Olawande Daramola

    cs.AI · cs.LG

    The advent of Large Language Models (LLMs) across many application domains has increased the need for Retrieval-Augmented Generation (RAG) to reduce hallucinations and improve LLM performance. Thus far, several RAG techniques with impressive features and capabilities have been proposed in the literature, making it a herculean task to select the most appropriate RAG techniques for specific problem scenarios. Thus, we propose a conceptual...

    arxiv.org/abs/2610.23226 · PDF

  32. 32

    Not All Test-Time Compute Is Evidence: Model-Specific Bias in LLM Reasoning

    Ming Shen, Tianjun Yao

    cs.AI

    Test-time scaling methods implicitly assume that additional computation from the same model provides independent evidence. We show that this assumption fails within a single reasoning model: errors are correlated across repeated samples, across reasoning depth, and in the model's own self-evaluation, so more compute from the same model is not more evidence. We introduce DIA (Disagreement-triggered Independent Arbitration), which treats...

    arxiv.org/abs/2610.23213 · PDF

  33. 33

    FuzzInsight: Navigating the Efficiency-Effectiveness Trade-off in Coverage-Guided DNN Fuzzing

    Vivek Vekariya, Fathiyeh Faghih, Alexander Pretschner

    cs.AI

    Collecting and labeling diverse test datasets for evaluating Deep Neural Networks (DNNs) is costly and time-consuming. This has motivated coverage-guided fuzzing techniques that generate defect-revealing corner cases from limited test data. However, their effectiveness depends heavily on configuration choices, particularly coverage criteria and seed-selection strategies. We present a configuration-oriented analysis and decision guidelines for...

    arxiv.org/abs/2610.23195 · PDF

  34. 34

    OfficeTown: Generative Organizations for Evolving Agent Intelligence

    Hongchen Wei, Daohong Ying, Zhi Hong, Aidan Wu, Bei Liu, Ruichun Ma, Yifan Yang, Qi Dai, Kai Qiu, Rui Xu, Dongdong...

    cs.AI

    AI agents increasingly use files and software tools to produce professional deliverables. Such work draws on materials, decisions, and feedback accumulated across assignments. We introduce OfficeTown, a generative simulation framework for professional organizations in which tasks and their context arise from continuing organizational work. From a natural-language description of an organization's purpose and activities, OfficeTown initializes...

    arxiv.org/abs/2610.23194 · PDF

  35. 35

    LongForge: Measuring and Improving Long-Context Instruction Data via Evidence-Assertion Reasoning Graphs

    Ziyang Chen, Junlong Jia, Xing Wu, Tinghao Yu, Songlin Hu

    cs.AI

    Existing quality control for long-context instruction-tuning data is a one-shot judgment inside the generation pipeline that leaves only a scalar, revealing neither which document parts an answer uses nor how many reasoning steps it writes down; the data can thus be neither audited nor controlled. We formalize data quality as a property of the alignment structure among document, question, and answer, and build LongForge, which pairs an...

    arxiv.org/abs/2610.23167 · PDF

  36. 36

    Algorithm Modelization: Transforming Optimization Algorithms into Evolvable Intelligence

    Jian Yao, Zhiyao Zhang, Yichen Li, Xingyu Wu, Kay Chen Tan

    cs.AI · cs.NE

    Can algorithms be learned and evolved as models, rather than repeatedly redesigned as programs? We explore this question by introducing Algorithm Modelization, a new paradigm that turns iterative algorithms into trainable models. Its central idea is to regard algorithm execution trajectories as computational experience: by learning how algorithmic states evolve over iterations, a model can internalize the knowledge embodied in existing...

    arxiv.org/abs/2610.23161 · PDF

  37. 37

    QuAD: Quantitative Assessment via Distribution-Alignment for Qualitative Data Quality Enabled by a Content-Aware Multi-Agent System

    Wenshuo Zhang, Leixian Shen, Qiqi Zhou, Shuchang Xu, Feiyang Hao, Huamin Qu, Haobo Li, Zhao Chen, Lin-Ping Yuan

    cs.AI · cs.HC

    Assessing whether qualitative responses collected from semi-structured questions are ready for a particular analysis goal is a persistent challenge. Manual qualitative analysis is interpretive and labor-intensive, while existing automated methods often lack a dataset-level perspective for deciding whether a response set supports a researcher-selected analytical lens. We introduce QuAD, a framework for quantitative pre-analysis assessment of...

    arxiv.org/abs/2610.23153 · PDF

  38. 38

    CIIR - Class-Indexed Interface Representation Learning

    Doo Yon Kim, Christian Wallraven

    cs.AI · cs.LG

    Multimodal learning commonly assumes paired or synchronized observations, yet independently collected datasets rarely provide such correspondence. We introduce CIIR, a modular encoder--decoder framework that connects heterogeneous datasets through a class-indexed address space while preserving each dataset's native input format and output module. A known local index selects address $k$, and an input-dependent 128-dimensional address...

    arxiv.org/abs/2610.23151 · PDF

  39. 39

    LexPilot: Benchmarking the Proactive Interaction Ability of Legal LLMs with Laypeople in Chinese Civil Case Consultation

    Wenbin Zhang, ShangZhan Li, Bingxiang He, Ye He, Yuxin Zhou, Muzhen Cai, Qi Shi, Wanxiang Che

    cs.AI

    Large Language Models (LLMs) have greatly assisted legal professionals. However, our empirical studies reveal that, when faced with underspecified consultations from non-experts, current LLMs often produce irresponsible responses with limited utility despite lacking sufficient case details for definitive conclusions. To better assist laypeople, legal LLMs must identify focal points of dispute and proactively elicit crucial context from users....

    arxiv.org/abs/2610.23141 · PDF

  40. 40

    WebCoder: Scaling General Frontend Development with Artifact-Grounded Agentic Reinforcement Learning

    Zijun Dou, Zhen Yang, Xijun Liu, Dongzhuoran Zhou, Aohan Zeng, Jie Tang

    cs.AI

    Real-world frontend development requires agents to compose reproduction, generation, modification, and repair over an evolving application state. Learning a general policy for this process requires translating heterogeneous execution feedback into task-aligned rewards and effective credit assignment across long sequences of interdependent actions. We introduce WebCoder, a general frontend development agent trained with artifact-grounded...

    arxiv.org/abs/2610.23104 · PDF

  41. 41

    Rethinking GRPO for Large Reasoning Model Safety Alignment

    Jiawei Chen, Yi Zhang, Jiawei Du, Zhonghou Lv, Tairan Chen, Junyuan Shang, Jinchang Hou, Xiao Yang, Yu Tian, Chenfu...

    cs.AI

    Large reasoning models (LRMs) exhibit strong reasoning capabilities but remain vulnerable to unsafe outputs. While Group Relative Policy Optimization (GRPO) is widely used for safety alignment, its group-relative optimization may provide insufficient safety-directed learning signals. In this paper, we identify two failure modes: advantage collapse, where homogeneous rollout groups yield vanishing advantages, and unsafe diversification, where...

    arxiv.org/abs/2610.23085 · PDF

  42. 42

    Agentic In-Context Learning: What Algorithms Can Transformers Learn in Closed-Loop Interaction?

    Shiguang Wu, Yaqing Wang, Quanming Yao

    cs.AI

    The learning algorithm view of in-context learning (ICL) studies how fixed-parameter Transformers learn from supplied examples. Language model agents extend this setting through interaction: their actions determine what evidence enters context next. We call this regime Agentic In-Context Learning (Agentic ICL). Existing studies largely depend on externally supplied contexts or reward-bearing interaction histories, while work on general...

    arxiv.org/abs/2610.23051 · PDF

  43. 43

    Distill Until Divergence: Error-Truncated Hindsight Distillation for Agentic RL

    Yibin Huang, Conghui Zhu

    cs.AI

    Long-horizon interaction requires agents to make multiple decisions in a changing environment. Outcome-based GRPO lacks fine-grained guidance for intermediate decisions. Recent methods distill completed trajectories into reusable skills and use hindsight-conditioned OPD to provide dense token-level credit. However, skill-induced probability shifts do not necessarily represent reward-aligned credit; especially after an erroneous action changes...

    arxiv.org/abs/2610.23048 · PDF

  44. 44

    Where Does the Performance Go? Understanding and Recovering Performance Regressions in C-to-Rust Translation

    Zhiqiang Yuan, Chong Wang, Zhenhao Zhou, Junwei Liu, Xin Peng

    cs.AI · cs.PL

    Rust is increasingly adopted as a memory-safe successor to C, and transpilers such as \crust{} can migrate entire C projects while preserving their behavior. Preserving behavior, however, does not guarantee preserving performance. We study this gap on 20 real-world C projects and their \crust{} translations. On the eight projects used for analysis, the translated Rust is more than 3\% slower than the original C on 12 of 19 operations, by up...

    arxiv.org/abs/2610.23044 · PDF

  45. 45

    Where Research Ideation Begins: Forward-Citation Supervision for Actionable Limitation Generation

    Liyang Zhao, Yong Cao, Haibin Ling

    cs.AI

    Automated research ideation has advanced rapidly, with multi-agent workflows that generate whole research ideas end to end. These frameworks leave largely unexamined the step that seeds a good idea: identifying an actionable limitation from literature. In this paper, we focus on this first step in the research ideation pipeline. Starting from a pilot study, we find that as large language models (LLMs) scale from 0.6B to 32B parameters, they...

    arxiv.org/abs/2610.23038 · PDF

  46. 46

    ASE: Attribution-Guided Semantic Evolution for Time Series Forecasting

    Yuxuan Zheng, Zhaoyang Li, Purui Liu, Hengbo Xiao, Guannan He

    cs.AI

    External text can supply predictive information absent from numerical histories, but its value depends on how it is represented, combined and deployed. We present Attribution-guided Semantic Evolution (ASE), a modular framework for constructing textual factors, evaluating their conditional contribution, and deploying validation-selected numerical--textual predictors. ASE supports task-specific implementations with lexical measurements,...

    arxiv.org/abs/2610.23027 · PDF

  47. 47

    Causal Value Attribution of Marketing Labels in LLM-Generated Ads: A Deconfounder Approach

    Danhong You, Wei Jiang, Zhiqiang, Zheng

    cs.AI · cs.CY

    As AI-generated content becomes commoditized, economic value increasingly shifts from final artifacts to the upstream expertise used to specify inputs that steer LLM generation. Yet organizations lack methods for valuing this expertise and identifying which individual skills or configurations causally improve performance. This study examines generative marketing, where structured expert labels serve as skill primitives conditioning...

    arxiv.org/abs/2610.23025 · PDF

  48. 48

    Adaptation of Computer Use Agents to New Software with User Manuals

    Chulhoon Yoon, Kyungri Park, Jinyeong Yun, Daeyun Jeong, Sukmin Yun, Woohwan Jung

    cs.AI

    Recent advances in multimodal foundation models have significantly improved the ability to operate real-world software through graphical user interfaces (GUIs) using visual observations and natural language instructions. However, maintaining strong performance on unseen applications remains challenging, as practical tasks often require long-horizon procedural knowledge and application-specific operation patterns. In addition, constructing...

    arxiv.org/abs/2610.23012 · PDF

  49. 49

    Efficient CPU-GPU collaborative MoE inference via Adaptive Expert Assignment

    Wenhui Lv, Yunming Liao, Hongli Xu, Yang Xu, Junhao Cheng

    cs.AI

    Mixture-of-Experts (MoE) models reduce inference computation through sparse expert activation, but their large expert memory footprint often exceeds GPU capacity, making expert offloading necessary for memory-constrained inference. However, existing systems typically handle GPU-missed experts by either transferring them to the GPU, incurring substantial PCIe transfer latency, or executing them on the CPU, making the slower CPU computation the...

    arxiv.org/abs/2610.23006 · PDF

  50. 50

    GOLD: Geometric Optimized Latent Diffusion for Structure-Aware RNA Inverse Folding

    Qi Si, Xuyang Liu, Penglei Wang, Shuo Su, Limei Han, Mengzhu Li, Xin Guo, Yuan Qi, Yuan Cheng

    cs.AI

    Designing RNA sequences that fold into specific 3D structures is a fundamental challenge in therapeutics. Current RNA inverse folding methods typically condition on 3D backbone coordinates but the generation is limited to 1D discrete sequences. This ignores the continuous side-chain orientations, which are essential for overall structural stability. To address this issue, we explicitly introduce side-chain geometry into the RNA generative...

    arxiv.org/abs/2610.22992 · PDF

  51. 51

    From Approval to Abuse: Privilege Escalation Attacks on Harness-Protected Coding Agents

    Xianglin Yang, Yechao Zhang, Jiahao Liu, Yulin Chen, Yuxin Cao, Bryan Hooi, Jiaheng Zhang

    cs.AI

    Coding agents use execution harnesses to constrain the consequences of prompt injection by enforcing permission boundaries. These systems deterministically intercept any out-of-sandbox action and prompt the user for approval. To reduce approval fatigue, users can approve allow-rules proposed by the agent, after which any future action matching the rule is allowed without a prompt. In this work, we show that this mechanism introduces a new...

    arxiv.org/abs/2610.22987 · 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.