AI news archive.
19,864 articles, newest first, from the outlets listed on the sources page.
- arXiv cs.AIresearchAccelerating Nonlinear MPC by Behavior Cloning with Separated Timescales: Application to Nuclear Reactor Load-Following
arXiv:2610.38496v1 Announce Type: new Abstract: Behavior Cloning (BC) of expert controllers for multiscale dynamical systems is challenging: state variables evolving at widely separated timescales induce highly nonlinear…
- arXiv cs.AIresearchValue Probes: Learned Value Functions for Beam-Search Red Teaming
arXiv:2610.38526v1 Announce Type: new Abstract: Automated red-team auditing of language models is an important way to evaluate safety-relevant behaviors. These audits implicitly search a tree of possible conversation pat…
- arXiv cs.AIresearchFrom Leniency to Suspicion: How an Open Post-Training Recipe Changes Whom LLM Fact-Checkers Believe
arXiv:2610.38532v1 Announce Type: new Abstract: LLMs increasingly fact-check online content with the poster's identity in view. However, it remains unclear which post-training changes how much a model's truth judgement d…
- arXiv cs.AIagentsAn Alignment Benchmark for Enterprise Agents on Era by Eon
arXiv:2610.38544v1 Announce Type: new Abstract: AI agents are widely used in enterprises to access business systems and act on them in pursuit of a goal. However, fulfilling a user's request can cause harm. Era by Eon is…
- arXiv cs.AIresearchPT$^3$: Prototype-Guided Self-Supervised Pretraining for Generalizable mmWave Human Point Cloud Representation
arXiv:2610.38551v1 Announce Type: new Abstract: Multi-source millimeter-wave (mmWave) observations offer an opportunity to learn human representations reusable across sensing tasks. However, sparse detections and multipa…
- arXiv cs.AIresearchCo-Evolving Policy and Rubric via Capability Awareness for Open-Ended Generation
arXiv:2610.38566v1 Announce Type: new Abstract: Rubric-based reinforcement learning (RL) offers interpretable, fine-grained supervision for aligning large language models (LLMs) on open-ended generation tasks. However, s…
- arXiv cs.AIresearchA-RSI: Adaptive Task Refresh for Zero-Data Multimodal Self-Improvement
arXiv:2610.38568v1 Announce Type: new Abstract: Recursive self-improvement requires models not only to acquire useful training experience, but also to decide when their current experience has ceased to be useful. Existin…
- arXiv cs.AIagentsCALM: Conflict-Aware Lifecycle Memory Governance for Long-Horizon Language Agents
arXiv:2610.38573v1 Announce Type: new Abstract: A memory can be outdated for present-day reasoning yet indispensable for reconstructing the past. Relevance-based retrieval overlooks this tension: semantic relevance, evid…
- arXiv cs.AIagentsThink Before You Act: Decision Gate for LLM Agents
arXiv:2610.38620v1 Announce Type: new Abstract: LLM agents increasingly perform consequential actions through external tools, yet their proposed actions are often executed without independent verification of policy compl…
- arXiv cs.AIagentsRight Numbers, Wrong Questions: Evaluating LLMs on Underspecified Financial Queries
arXiv:2610.38631v1 Announce Type: new Abstract: Financial assistants answer questions that sound factual but depend on what the user leaves unsaid: what a data figure measures, which product variant is meant, and the inv…
- arXiv cs.AIresearch"Close enough?" An investigation into multiplication errors in LLMs
arXiv:2610.38634v1 Announce Type: new Abstract: Large language models (LLMs) frequently fail at multi-digit arithmetic, but the underlying mechanisms governing these failures remain poorly understood. While recent work d…
- arXiv cs.AIagentsCurriculum-Based Deep Reinforcement Learning for Robust Autonomous Rocket Landing under Atmospheric Uncertainty
arXiv:2610.38636v1 Announce Type: new Abstract: Autonomous far-range rocket landing is a challenging control problem because navigation, descent-rate regulation, attitude management, and fuel use must be coordinated in t…
- arXiv cs.AIresearchWeaken to Strengthen: Weakening-Induced On-Policy Self-Distillation for Large Reasoning Models
arXiv:2610.38681v1 Announce Type: new Abstract: Self-distillation offers a way to improve reasoning models when a suitable external teacher is unavailable. However, common approaches rely on privileged information to con…
- arXiv cs.AIresearchDeogen: Automatic Benchmark Generation for Reasoning over Policies
arXiv:2610.38687v1 Announce Type: new Abstract: Language models are increasingly being deployed in regulated settings, where a model must apply given rules to the facts of a situation to determine a verdict. Benchmarking…
- arXiv cs.AIagentsBeyond fabrication: Misattributed numbers in papers written by AI research agents
arXiv:2610.38691v1 Announce Type: new Abstract: Verifying numerical claims in research written by AI agents is an important but challenging problem. Recent benchmarks show that many agent-written papers fail integrity ch…
- arXiv cs.AIresearchFrom Grind to Grasp
arXiv:2610.38728v1 Announce Type: new Abstract: Discussions of complexity, knowledge and intelligence often overlook the work involved. For example, outputs generated by short programs have low algorithmic complexity but…
- arXiv cs.AIresearchLeaving, Staying, or Living There? Duration Commonsense Reasoning with Human-LLM Co-annotations
arXiv:2610.38755v1 Announce Type: new Abstract: Gardening in the backyard implies living there, while checking into an Airbnb more likely means staying for a few days. Such duration judgments are important for understand…
- arXiv cs.AIresearchLearning to Reason from Reference Samples Alone: Reinforced Adversarial Learning for Modern LLM Post-Training
arXiv:2610.38851v2 Announce Type: replace Abstract: In many settings where LLM post-training is desirable, high-quality answers are recognizable but difficult to score with explicit rewards. Reinforcement learning with v…
- arXiv cs.AIagentsWhere You Inject Matters: The Trust Region Effect in LLM Agents
arXiv:2610.38867v1 Announce Type: new Abstract: LLM agents are increasingly adopted in long-horizon tasks and interact with diverse external environments. Despite their growing capabilities, these agents remain vulnerabl…
- arXiv cs.AIagentsAuditable Belief Updates as a Missing Safety Layer for LLM-Powered Multi-Agent Systems
arXiv:2610.38869v1 Announce Type: new Abstract: Large Language Models (LLMs) are increasingly being used as cognitive interfaces that shape how users interpret evidence, communicate, and make decisions. As deployed syste…
- arXiv cs.AIagentsAligning Autoresearch: Co-Improvement between Agents and Humans
arXiv:2610.38882v1 Announce Type: new Abstract: Autoresearch agents can now attempt long-horizon AI research end-to-end, i.e., self-improvement. However, best results are still obtained by research agents interacting wit…
- arXiv cs.AIagentsTreasuryQA Bench: Can AI Agents Work an Archive of Scanned Financial Records?
arXiv:2610.38924v1 Announce Type: new Abstract: Organisations keep decades of financial records as scanned PDFs on shared drives, and AI agents can now reach those drives through Model Context Protocol (MCP) servers. Exi…
- arXiv cs.AIagentsEfficient Agentic Audio-Visual Question Answering through Toolset Pruning and Modality Consolidation
arXiv:2610.38960v1 Announce Type: new Abstract: Fine-grained audio-visual question answering (AVQA) requires models to resolve complex cross-modal correspondences and temporally grounding from sparsely distributed eviden…
- arXiv cs.AIresearchDungeons & Dragons Encounter Analysis with Semi-Random Simulation in Foundry VTT
arXiv:2610.38975v1 Announce Type: new Abstract: Dungeons & Dragons (D&D) is a tabletop roleplaying game (TTRPG) in which a group of players, guided by a Game Master, roleplay a fantasy adventure with simulated combat and…
- arXiv cs.AIresearchContext Compaction for World Models
arXiv:2610.38980v1 Announce Type: new Abstract: World models let observers explore and interact with generated worlds. Recent work shows that when observers revisit a region, current models often fail to regenerate it co…
- arXiv cs.AIresearchSyntactic Segmentation and Collage Generation for Videogame Levels with Perceptual Distance
arXiv:2610.38987v1 Announce Type: new Abstract: In procedural videogame level design, many existing techniques combine hand-authored level chunks. Some existing analysis techniques also rely on a higher-level graph struc…
- arXiv cs.AIagentsAgentic AI Reasoning and Learned Sampling for Task-Aware Projection Selection in Sparse-View CT: Predicting Hemodynamics
arXiv:2610.39003v1 Announce Type: new Abstract: Sparse-view computed tomography (CT) reduces the number of acquired projections, but conventional sampling strategies are designed primarily for image reconstruction rather…
- arXiv cs.AIresearchFormal Specifications with Data-Driven Signal Analysis
arXiv:2610.39007v1 Announce Type: new Abstract: Developing analog and hybrid systems requires substantial domain knowledge that is hard to articulate in a formal specification; without additional effort, this domain know…
- arXiv cs.AIresearchWhen Should AI Decide, Defer, and Deviate? Responsible Automation for Deceased Donor Kidney Allocation
arXiv:2610.39045v1 Announce Type: new Abstract: We ask when an AI trained on historical decisions should be allowed to recommend decisions, defer to human experts, or in other cases, flag concerning patterns in data and…
- arXiv cs.AIresearchWhen the Same Prompt Gives Different Answers (and the Same Answer Isn't Always Right): A Taxonomy of Failure Mechanisms, Testing, and Mitigation Strategies for LLM-Assisted Evidence Synthesis Read-and-Evaluate Tasks
arXiv:2610.39070v1 Announce Type: new Abstract: LLM read-and-evaluate tasks include study screening, data extraction, risk-of-bias assessment, and critical appraisal in the evidence synthesis field. These task differ in…
- arXiv cs.AIresearchCertified Access: Test-Time Compute on Conservative Physics-Structured AI Memory
arXiv:2610.39073v1 Announce Type: new Abstract: Test-time compute mechanisms, such as retries, confidence-based escalation and non-local routing, are conventionally deployed as black-box additions to trained models. They…
- arXiv cs.AIresearchReasoning or Pattern Exploitation? Mechanistic Insights from RL-Trained Language Models
arXiv:2610.39182v1 Announce Type: new Abstract: Large language models (LLMs) are increasingly described as acquiring "reasoning" skills after reinforcement learning with verifiable rewards. However, whether these gains r…
- arXiv cs.AIagentsGoal-Aware Dynamic Multi-Agent Collaboration for Long-Horizon Tasks
arXiv:2610.39193v1 Announce Type: new Abstract: LLM-based agents are increasingly used to complete complex, long-horizon tasks without human intervention. To make progress toward the task goal, agents need to perform dif…
- arXiv cs.AIresearchBeyond LLM Wrappers: Context Engineering and the Three-Engine Architecture for Reliable Financial AI
arXiv:2610.39230v1 Announce Type: new Abstract: Financial AI systems must support decision-bearing questions involving payments, collections, cash flow, and compliance while ensuring numerical accuracy, traceability, and…
- arXiv cs.AIresearchSocialArena: Evaluating Social Intelligence in Large Language Models
arXiv:2610.39247v1 Announce Type: new Abstract: We introduce SocialArena, a benchmark for systematically defining and evaluating the social intelligence of large language models (LLMs). Although LLMs are increasingly dep…
- arXiv cs.AIagentsWhere to Learn and How Far to Roll Out: Efficient Adaptive Training for GUI Agents
arXiv:2610.39251v1 Announce Type: new Abstract: Trajectory-level GRPO is a common approach for training multi-step GUI agents, but repeatedly rolling out complete trajectories from the task root can be inefficient: task-…
- arXiv cs.AIagentsOnline Dense Reinforcement Learning with Planning-Grounding Agent for UI Navigation on Live IOS Apps
arXiv:2610.39264v1 Announce Type: new Abstract: While frontier vision-language models demonstrate strong capabilities in user interface (UI) navigation, their high inference costs prohibit everyday use. Conversely, small…
- arXiv cs.AIresearchLLM-Assisted Optimization for Simulation-Based Redesign
arXiv:2610.39275v1 Announce Type: new Abstract: Engineering design frequently involves redesigning a prior solution to adapt to changes in context, such as material prices or system requirements. Redesign often leverages…
- arXiv cs.AIagentsMemSteer: Turning User Histories into Supervision for Agent Memory Optimization
arXiv:2610.39276v1 Announce Type: new Abstract: An effective agent memory system should adapt to a user's history and task requirements. Human-annotated supervision can guide this adaptation by providing feedback for eva…
- arXiv cs.AIagentsWhat Does "Now" Mean to Your Computer-Use Agent? Stale Screenshots, Page Loads, and Which Check to Run Before a Click
arXiv:2610.39282v1 Announce Type: new Abstract: Computer-use agents plan on a screenshot for several seconds and then click at a position on the screen; if the page has changed in the meantime, the click lands on an unin…
- arXiv cs.AIresearchProving and Improving the Platonic Representation Hypothesis with JEPAs
arXiv:2610.39302v1 Announce Type: new Abstract: Whether models trained on different modalities converge to the same representation of the world is central to AI. The Platonic Representation Hypothesis (PRH) conjectures t…
- arXiv cs.AIresearchCoToan: Executable Rules and a Representation Contract for Reproducible Game AI Evaluation
arXiv:2610.39353v1 Announce Type: new Abstract: Game AI results are comparable only when different implementations expose the same decision problem. Yet seemingly minor representation choices can silently change that pro…
- arXiv cs.AIagentsBefore the Next Rollout: Diagnosing Skill Utility Shifts from Policy Updates
arXiv:2610.39358v1 Announce Type: new Abstract: Skill-evolution methods typically analyze execution trajectories to guide agent skill refinement. However, policy updates can change the behavioral effects and marginal uti…
- arXiv cs.AIagentsAgentDecept: Benchmarking Deceptive Behavior in Tool-Using LLM Agents under Pressure
arXiv:2610.39377v1 Announce Type: new Abstract: LLM agents increasingly carry out multi-turn tasks across everyday applications, yet users often see only the agent's report rather than the observations, API calls, and st…
- arXiv cs.AIresearchRetaining Repair Hypotheses for Cost-Aware Execution
arXiv:2610.39410v1 Announce Type: new Abstract: Several corrections of an executable model can fit the same observations while predicting different execution outcomes. We study a repair-to-execution interface that retain…
- arXiv cs.AIresearchVerifiable LLM-Guided Focal SMT Solving for Quantified Arrays
arXiv:2610.39435v1 Announce Type: new Abstract: Large language models can often identify useful semantic structure in formal artifacts, but their outputs cannot be trusted as proofs or solver decisions. We study whether…
- arXiv cs.AIagentsATLAS-Bench: An Agentic Tool-Use Benchmark for Local Area Search
arXiv:2610.39438v1 Announce Type: new Abstract: Reliable local search is essential for LLM agents to support everyday decisions and longer plans by finding places that meet users' needs at given times and locations. Yet…
- arXiv cs.AIresearchTimbre: Reliability-aware harmonic consensus for label-free multimodal test-time rexinforcement learning
arXiv:2610.39439v1 Announce Type: new Abstract: Label-free test-time reinforcement learning adapts vision--language models using rewards derived from their own predictions. However, misinterpreted visual evidence and sen…
- arXiv cs.AIresearchKairos: Semantics-Aware Orchestration of Planning Engines across Classical, Numeric, Temporal and Hybrid PDDL
arXiv:2610.39454v1 Announce Type: new Abstract: Planning at run time, as in robots that re-plan while they act, needs a planner that reads the PDDL its users write (durative actions, numeric resources, time windows, cont…
- arXiv cs.AIresearchDoes Federated Training Degrade Explainability? A Comparative Analysis of Grad-CAM Faithfulness in Centralized vs. Federated Deepfake Detection
arXiv:2610.39483v1 Announce Type: new Abstract: Automated detection has become the essential safeguard against realistic synthetic media, but deploying it at scale requires training on sensitive data from multiple organi…