AI news archive.
19,864 articles, newest first, from the outlets listed on the sources page.
- arXiv cs.AIresearchFrom Unreliable to Usable: Decomposing Failure Modes and Confidence Signals in LLM Entity Extraction - Evidence from Cyber Threat Intelligence
arXiv:2610.20417v1 Announce Type: cross Abstract: Audits of large language model (LLM) extraction from full-length documents report severe unreliability, leaving two questions open: what the failures actually consist of,…
- arXiv cs.AIresearchOrionMoE-v1.0: A Mixture-of-Experts Foundation Model for Tabular In-Context Learning
arXiv:2610.20418v2 Announce Type: replace-cross Abstract: Tabular foundation models enable in-context learning by predicting unseen observations from a labeled support table without task-specific parameter updates. Despi…
- arXiv cs.AIresearchAnalyzing performance of Retrieval Augmented Generation models for Question Answering Tasks
arXiv:2610.20419v2 Announce Type: replace-cross Abstract: Retrieval-Augmented Generation (RAG) models have become almost a cornerstone of open-domain question answering tasks; in which the quality of answers generated in…
- arXiv cs.AIagentsFrom Training to Escalation: Forensic Analysis and Local Recreation of RL/RLHF-Induced Privilege Escalation in AI Agents (CVE-2026-65616)
arXiv:2610.20437v1 Announce Type: cross Abstract: Mid-2026 reporting on autonomous AI evaluation incidents claimed that frontier large language model (LLM) agents escaped server containment through emergent coordination…
- arXiv cs.AIresearchEqual CER, Unequal Damage: Severity-Weighted Character Error Rate for OCR
arXiv:2610.20461v1 Announce Type: cross Abstract: Character Error Rate (CER), the standard metric for OCR and vision-language models (VLMs), charges a typo as much as a wrong name. We introduce semCER, which keeps CER's…
- arXiv cs.AIresearchSelective Growing Q-Networks (SGQN): An Analysis of Adaptive Action Grids in Decoupled Q-Learning
arXiv:2610.20463v1 Announce Type: cross Abstract: Value-based Q-learning for continuous control can discretise the action space, but the resolution of the discretisation trades early exploration against final precision.…
- arXiv cs.AIresearchQuantifying privacy risks in foundation model representations of brain MRI
arXiv:2610.20494v1 Announce Type: cross Abstract: Pretrained representations are increasingly used to share and reuse information from medical images without exposing the original scans, but it is unclear how much patien…
- arXiv cs.AIresearchHindsight-Privileged RLVR: Single-Rollout Token Credit Without Group Sampling
arXiv:2610.20497v2 Announce Type: replace-cross Abstract: Reinforcement learning with verifiable rewards (RLVR) trains language models from checked task outcomes, but assigning a terminal reward to individual reasoning t…
- arXiv cs.AIresearchVascuSR: Iterative Anatomy-Aware Diffusion for CT Super-Resolution
arXiv:2610.20509v1 Announce Type: cross Abstract: Generative super-resolution (SR) can recover sharp detail in computed tomography (CT) images from severely downsampled observations, but the missing high-frequency conten…
- arXiv cs.AIresearchVASPO: Self-Distillation with Privileged Value Estimation
arXiv:2610.20516v1 Announce Type: cross Abstract: On-policy self-distillation provides dense supervision for reasoning models via a privileged self-teacher. However, this supervision reflects local token preferences rath…
- arXiv cs.AIresearchPrismaCoder: Pushing the Boundaries of Open Multimodal Code Intelligence through Decomposed Rubric Synthesis
arXiv:2610.20538v1 Announce Type: cross Abstract: Multimodal code intelligence connects the symbolic world of programs with the visual world their execution creates. Interest in this capability is burgeoning, as models a…
- arXiv cs.AIresearchDetection of Potato Plant Leaf Diseases using Convolutional Neural Networks
arXiv:2610.20544v1 Announce Type: cross Abstract: Potato is an important food-security crop in Ethiopia, but its production is affected by foliar diseases, particularly early blight and late blight. This study develops a…
- arXiv cs.AIagentsPerSo: Investigating Facets of LLMs' Social Intelligence in Real-World Persuasion Scenarios
arXiv:2610.20548v1 Announce Type: cross Abstract: Understanding the social intelligence of large language models (LLMs) is increasingly important for applications involving human-AI interaction and multi-agent environmen…
- arXiv cs.AIresearchRecovering Reliable Supervision via Critique-Guided On-Policy Distillation
arXiv:2610.20572v1 Announce Type: cross Abstract: On-policy distillation (OPD) trains a student model on its own rollouts with token-level supervision from a teacher, assuming that the teacher can reach a correct answer…
- arXiv cs.AIagentsBeyond LLM Judge: Mechanical Verification for Agentic Autoformalization
arXiv:2610.20573v2 Announce Type: replace-cross Abstract: We introduce a novel method for verifying Lean formalizations that goes beyond LLM-judged faithfulness. Unlike most verification methods, which rely on LLMs or tr…
- arXiv cs.AIagentsOne Task, Many Interfaces: Simple Cross-Interface Consistency Training Improves LLM Agent Robustness
arXiv:2610.20576v1 Announce Type: cross Abstract: Post-training binds the competence of an LLM agent to the interface it was trained on. Semantics-preserving perturbations of that interface, such as renaming an action or…
- arXiv cs.AIagentsPlayer-1: Learning to Play from Synthetic Experience
arXiv:2610.20584v1 Announce Type: cross Abstract: Computer-use agents have demonstrated strong capabilities in interpreting screens and executing keyboard and mouse actions. However, games, with their visually rich world…
- arXiv cs.AIresearchContinuum Attention
arXiv:2610.20586v2 Announce Type: replace-cross Abstract: Conventional causal attention aggregates values at discrete memory positions. We introduce Continuum Attention (CA), in which neighboring key and value vectors de…
- arXiv cs.AIagentsPrice of Anarchy of First-Price Auctions in Agentic Markets
arXiv:2610.20587v1 Announce Type: cross Abstract: An agentic market is a decentralized serving-time market where AI agents bid to claim each user prompt. An agent's bid is based on its expected value for the prompt, whic…
- arXiv cs.AIresearchYour Transformer Is Secretly Elastic
arXiv:2610.20588v1 Announce Type: cross Abstract: Scaling up transformers has produced models with broad capabilities, deployed across heterogeneous hardware and fluctuating workloads. Yet, model families offer only a fe…
- arXiv cs.AIresearchModeling and Understanding Neural Weights of Humanoid Policies with Diffusion
arXiv:2610.20606v1 Announce Type: cross Abstract: Recent advances in robot learning have driven rapid progress in humanoid control, enabling increasingly agile behaviors in the real world. However, humanoid control netwo…
- arXiv cs.AIagentsHierarchical Residual Policy Learning for Real-World Mobile Manipulation with Sparse Human Guidance
arXiv:2610.20607v1 Announce Type: cross Abstract: Mobile Manipulation (MoMa) is a cornerstone of autonomous robotic systems, yet real-world fine-tuning remains a formidable challenge because of high-dimensional action sp…
- arXiv cs.AIagentsPrompt Stability Matters: Evaluating and Optimizing Auto-Generated Prompt in General-Purpose Systems
arXiv:2505.13546v2 Announce Type: replace Abstract: Automatic prompt generation plays a crucial role in enabling general-purpose multi-agent systems to perform diverse tasks autonomously. Existing methods typically evalu…
- arXiv cs.AIagentsSEDM: Selective Self-Evolving Distributed Memory for Agents
arXiv:2509.09498v4 Announce Type: replace Abstract: Agents in a society can reuse each other's experience, but indiscriminate memory use increases processing cost and can reduce task success. We present SEDM (Selective S…
- arXiv cs.AIresearchDual-Evolving Reinforcement Learning for Automated Reward Design
arXiv:2512.13399v4 Announce Type: replace Abstract: Crafting effective reward signals remains a central challenge in reinforcement learning (RL) for complex reasoning. Existing methods for automated reward optimization s…
- arXiv cs.AIagentsEmergence of Physical Intelligence via Controllable Information Production
arXiv:2601.22449v3 Announce Type: replace Abstract: Intrinsic Motivation (IM) aims to train agents without external rewards, enabling useful behavior to emerge from the agent's interaction with its environment alone. How…
- arXiv cs.AIresearchLingLanMedBench: A Unified Benchmark for Evaluating Large Language Models in Traditional Chinese Medicine
arXiv:2602.01779v2 Announce Type: replace Abstract: Reliable evaluation of large language models in traditional Chinese medicine remains limited because existing benchmarks mainly focus on examination-style question answ…
- arXiv cs.AIresearchTKGB: A Comprehensive Benchmark for Temporal Knowledge Graph Forecasting
arXiv:2602.08353v2 Announce Type: replace Abstract: Temporal knowledge graph (TKG) forecasting has advanced remarkably in recent years. Yet we found existing benchmarks inadvertently introduce a shortcut: For a subset of…
- arXiv cs.AIresearchDraft-Thinking: Learning Efficient Reasoning in Long Chain-of-Thought LLMs
arXiv:2603.00578v2 Announce Type: replace Abstract: Long chain-of-thought~(CoT) reasoning improves the problem-solving capability of large reasoning models~(LRMs), but often incurs substantial inference cost and overthin…
- arXiv cs.AIagentsTimeWarp: Evaluating Web Agents by Revisiting the Past
arXiv:2603.04949v4 Announce Type: replace Abstract: As web agents close the gap with humans on benchmarks, one question arises: Do today's agents perform just as well on tomorrow's web? We introduce TimeWarp, a benchmark…
- arXiv cs.AIresearchThe Reasoning Trap -- Logical Reasoning as a Mechanistic Pathway to Situational Awareness
arXiv:2603.09200v2 Announce Type: replace Abstract: Situational awareness, the capacity of an AI system to recognize its own nature, understand its training and deployment context, and reason strategically about its circ…
- arXiv cs.AIresearchBenchPreS: A Benchmark for Context-Aware Personalized Preference Selectivity of Persistent-Memory LLMs
arXiv:2603.16557v3 Announce Type: replace Abstract: Large language models (LLMs) increasingly store user preferences in persistent memory to support personalization across interactions. However, in third-party communicat…
- arXiv cs.AIresearchLearning to Disprove: Formal Counterexample Generation with Large Language Models
arXiv:2603.19514v2 Announce Type: replace Abstract: Mathematical reasoning demands two critical, complementary skills: constructing rigorous proofs for true statements and discovering counterexamples that disprove false…
- arXiv cs.AIresearchReasoning or Rhetoric? An Empirical Analysis of Moral Reasoning Explanations in Large Language Models
arXiv:2603.21854v2 Announce Type: replace Abstract: Do large language models reason morally, or do they merely sound like they do? We investigate whether LLM responses to moral dilemmas exhibit genuine developmental prog…
- arXiv cs.AIresearchStaRPO: Stability-Augmented Reinforcement Policy Optimization
arXiv:2604.08905v2 Announce Type: replace Abstract: Reinforcement learning (RL) is effective in enhancing the accuracy of large language models in complex reasoning tasks. Existing RL policy optimization frameworks rely…
- arXiv cs.AIresearchReward Design for Physical Reasoning in Vision-Language Models
arXiv:2604.13993v2 Announce Type: replace Abstract: Physical reasoning over images requires integrating perception, domain knowledge, and multi-step symbolic inference, yet Vision-Language Models (VLMs) fall far short of…
- arXiv cs.AIresearchDiffTSP: A Conditional Discrete Diffusion Model for Knowledge Graph Triple Set Prediction
arXiv:2604.18344v2 Announce Type: replace Abstract: Triple set prediction (TSP) aims to discover missing facts from an incomplete knowledge graph without predefined queries. Existing approaches typically construct the pr…
- arXiv cs.AIresearchIdeological Bias in LLMs' Economic Causal Reasoning
arXiv:2604.21334v3 Announce Type: replace Abstract: Do large language models (LLMs) exhibit systematic ideological bias when reasoning about economic causal effects? As LLMs are increasingly used in policy analysis and e…
- arXiv cs.AIagentsTowards Security-Auditable LLM Agents: A Unified Graph Representation
arXiv:2605.06812v2 Announce Type: replace Abstract: LLM-based agentic systems are rapidly evolving to perform complex autonomous tasks through dynamic tool invocation, stateful memory management, and multi-agent collabor…
- arXiv cs.AIresearchCORTEG: Foundation Models Enable Cross-Modality Representation Transfer from Scalp to Intracranial Brain Recordings
arXiv:2605.10337v2 Announce Type: replace Abstract: Intracranial electrocorticography (ECoG) offers high signal-to-noise access to cortical activity for brain-computer interfaces, yet limited per-patient data has led mos…
- arXiv cs.AIagentsHow Far Are We From True Auto-Research?
arXiv:2605.19156v2 Announce Type: replace Abstract: Recent auto-research systems can produce complete papers, but feasibility is not the same as quality, and the field still lacks a systematic study of how good agent-gen…
- arXiv cs.AIagentsMOCHA: Multi-Objective Chebyshev Annealing for Agent Skill Optimization
arXiv:2605.19330v2 Announce Type: replace Abstract: LLM agents organize behavior through skills -- structured natural-language specifications that govern how an agent reasons, retrieves, and responds. Unlike monolithic p…
- arXiv cs.AIresearchC-MIG: Multi-view Information Gain-based Retrieval-Augmented Generation for Clinical Diagnosis Reasoning
arXiv:2605.27860v3 Announce Type: replace Abstract: Retrieval-augmented generation (RAG) trained with reinforcement learning (RL) is a promising way to ground large language models in medical evidence for clinical diagno…
- arXiv cs.AIagentsRedundant or Necessary? A Benchmark for Detecting Redundant Steps in Agent Trajectories
arXiv:2605.29893v2 Announce Type: replace Abstract: Computation cost is a critical challenge for LLM agent systems, as their long-horizon trajectories continuously accumulate computational overhead. While existing studie…
- arXiv cs.AIagentsLearning Reusable Agent Subroutines from Experience
arXiv:2606.02994v2 Announce Type: replace Abstract: Existing methods induce various forms of knowledge from agent traces, from sample workflows to programmatic skills. We propose ARISE (Agent Routine Induction from Succe…
- arXiv cs.AIresearchAUDITFLOW: Executable Symbolic Environments for Structured Financial Reporting Verification
arXiv:2606.03031v2 Announce Type: replace Abstract: Structured financial reporting verification requires connecting reported facts to accounting constraints, reconstructing expected values, and executing numerical checks…
- arXiv cs.AIresearchPosition: Hippocampal Explicit Memory Is the Cornerstone for AGI
arXiv:2606.11245v2 Announce Type: replace Abstract: Large Language Models (LLMs) have demonstrated remarkable capabilities across various tasks, raising expectations for Artificial General Intelligence (AGI). This positi…
- arXiv cs.AIagentsCausalDS: Benchmarking Causal Reasoning in Data-Science Agents
arXiv:2607.08093v2 Announce Type: replace Abstract: Large language models (LLMs) increasingly act as data-science agents, combining abstract reasoning with advanced tool use. Yet the relevant benchmark landscape largely…
- arXiv cs.AIresearchMaking Transformers Logical
arXiv:2608.00418v2 Announce Type: replace Abstract: Transformers do not merely lack data on some Boolean extrapolation tasks; they generalize in a systematically wrong way. Recent work on generalization on the unseen has…
- arXiv cs.AIagentsWhen Privileged Guidance Misaligns: State-Matched Routing and Contextualized Self-Distillation for Multi-Turn Agents
arXiv:2608.05219v2 Announce Type: replace Abstract: Multi-turn agents trained with sparse outcome rewards receive limited supervision for intermediate decisions. Privileged on-policy self-distillation provides denser gui…