AI News

Upstream Influence, Downstream Re-Anchoring in LLM Agent Chains (arxiv.org)

Evidence-First Agentic Verification (arxiv.org)

Safety for Whom? Benchmarking Understanding Beyond Refusal in Egocentric Multimodal Assistants (arxiv.org)

DuplexAct: How Interactive Are Frontier Full-Duplex Spoken Language Models? (arxiv.org)

Coupled Harness--Model Evolution for Scientific Equation Discovery (arxiv.org)

Instructions Are Evidence, Not Objectives: Measuring Outcome Alignment of Agents (arxiv.org)

Autonomous 3D Visual-Thermal Inspection for Utility-Scale Solar Farms (arxiv.org)

Memory Echo in Single-Agent Long-Term Memory: Lifecycle Evaluation and Provenance-Guided Mitigation (arxiv.org)

IsoVRBench: Video Reward Modeling Without Human Feedback (arxiv.org)

Memory Without Curation: Learning Multi-Stream Latent Memory for Robot Policies (arxiv.org)

Coarse-Graining Recursive Expert Computation: Continuation-Relative Interfaces and Failure Modes (arxiv.org)

SceneForesee: Learning Scene Evolution for Spatiotemporal Reasoning (arxiv.org)

Target-ESS Tempered Policy Optimization (TETPO): Smooth Variance Control via Bridge Importance Sampling (arxiv.org)

Specification-Based Testing of Large Language Models (arxiv.org)

KV4-RL: Unlocking Faster Asynchronous Rollouts for RLVR with INT4 KV Cache Quantization (arxiv.org)

Jevolution: A Framework for Studying Communication and Collective Dynamics in AI Agents (arxiv.org)

Co-adaptive Rule Routing and Constraining for Structured Knowledge Retrieval (arxiv.org)

Diffusion-Based Auctions for Multi-Winner Banner Ads (arxiv.org)

RES-MR: Risk-Aware Reasoning for Explainable and Safe Medication Recommendation (arxiv.org)

Scale, Drivers, and Limits of Representational Alignment Across Large Language Models (arxiv.org)

What Counts as the Same Task? Executable Contracts for Robustness Evaluation of Tool-Using Agents (arxiv.org)

AphaVoice: Text-to-Speech Model for Aphasia Patient Simulation (arxiv.org)

Rethinking Activation Functions: Reducing Bias Can Improve Learning (arxiv.org)

Compositional Concepts Erasing: Disentangling Object-Style via Domain Manifold Unlearning (arxiv.org)

Who to Trust in Multi-Robot Systems: Joint Robot and Object Hypothesis Trust Inference via Neural-Symbolic Cross-Validation (arxiv.org)

GG-Nav: Goal-oriented Grounding Chain-of-Thought Elicits Reasoning in Navigation Foundation Models (arxiv.org)

A Self-Defining Operator for Distributed Applications (arxiv.org)

Active Neural Matter: A Finite-Window Response Framework for Physical Systems and Distributed AI Workflows (arxiv.org)

Learning Adaptive Deliberation Control for Reliable Recognition with Multi-Agent Language Models (arxiv.org)

Socratic Skill Learning: Interleaving Teaching and Practice for Transferable Agent Skills (arxiv.org)

ECMR: Evidence-Aware Deliberation Control for Multi-Agent LLM Reasoning (arxiv.org)

Training LLMs for Multi-Agent Collaboration with Verifiable Information Rewards (arxiv.org)

Seeing Depth but Choosing Cues in Vision-Language Models (arxiv.org)

SHELDON: Reasoning through Dual Search over Hypothesis and Observation Spaces (arxiv.org)

Can Agents Forget When Asked? A Benchmark for Memory Unlearning in LLM Agents (arxiv.org)

WorkRover: Scaling Long-Horizon Cowork Agent Training Via Recursive Environment Evolution (arxiv.org)

Two Routes to Interoceptive Prioritisation: Dissociating Likelihood Precision and Preference Reweighting in Active Inference (arxiv.org)

AReaL-RSI: Adaptive Scaling and Coordination of Recursive Self-Improvement (arxiv.org)

Chain of Thought Monitoring Can Fail Even When CoT Is Necessary (arxiv.org)

One-Sided Verification Under Limited Expert Review (arxiv.org)

AM-Bench: Understanding Agents under Task Underspecification across Models and Harnesses (arxiv.org)

LifeSearch: Benchmarking Deep Research Agents for Real-World Life and Productivity (arxiv.org)

CompanionSafe: From Multi-turn Safety Evaluation to Guardrails for AI Companions (arxiv.org)

Fresh at Decision, Stale at Action: Routing New Information to LLM Decisions Already Made (arxiv.org)

Monitoring latent reasoning in looped language models (arxiv.org)

Mitigating Cross-Modal Hallucinations by Correcting Modality Misrouting in Omnimodal LLMs (arxiv.org)

RadixFlow: Filling Execution Bubbles in Prefix-Cached Hybrid LLM Serving (arxiv.org)

Beyond Answer Sharing: TrueMAD for Targeted Peer Engagement in Multi-Agent Debate (arxiv.org)

When and How Order-Flow Interventions Work After Dealers Adapt (arxiv.org)

Back to the Future: Reverse-Consequence Modeling for Long-Horizon Particle Dynamics (arxiv.org)