AI News

Infinity Machine: a self-improving AI agent for research (arxiv.org)

CANS: Capability-Preserving Neuron Steering for Large Language Models (arxiv.org)

Test-Time Confidence Adaptation under RL-Induced Trajectory Shift (arxiv.org)

ShutdownMod: Turning Any Trajectory into a Shutdown Evaluation (arxiv.org)

Retriever-Free Retrieval: Navigating Documents with Decision Models (arxiv.org)

EvoBench: An Evolving, Diagnostic Benchmark Framework for Environmental Modeling (arxiv.org)

Beyond Federated Learning: Decentralized Agentic Systems for Self-Evolving Clinical Intelligence (arxiv.org)

Omni-InteractScope: Benchmarking Proactive and Agentic Interaction in Continuous Omnimodal Streams (arxiv.org)

Skill$^{\infty}$: Recursive Self-Improvement for Agent Skill Optimization (arxiv.org)

CaliMem: Search-Efficient Data Agent via Calibrated Memory (arxiv.org)

What Effort Level Should I Choose? On the Economics of Reasoning Effort in LLM Agents (arxiv.org)

From Outranking Flow Scores to Linear Programs: Group Decision Making in PROMETHEE II (arxiv.org)

CTXAgent - Self-Controlled Context Management of LLM Agents (arxiv.org)

Matched Capabilities, Different Behaviors: A Controlled Study of MCP and CLI Interfaces for Language Agents (arxiv.org)

Intervention-Certified Amplification: A Validity Gate for $\sigma$-Weighted Post-Training (arxiv.org)

5W1H+Which: Context-Valid Semantic Indexing with Progressive Ontology Binding (arxiv.org)

JRDB-AVR: An Active Visual Reasoning Benchmark for Embodied Agents in Real-World Environments (arxiv.org)

EOPSA: Efficient On-Policy Self-Distilled Safety Alignment (arxiv.org)

A Computer Vision Approach to Visual Fraud Detection in Phishing Websites Using YOLOv8 (arxiv.org)

Hybrid-time PINNs: Discontinuity-Aware Physics-Informed Neural Networks for Learning Hybrid Dynamical Systems (arxiv.org)

DISCERN: Can AI Agents Work Like Scientists and Guide Discovery? (arxiv.org)

HARP-LoRA: Hierarchy-Conditioned Rank Projection for Low-Rank Adaptation in Scientific Table Understanding (arxiv.org)

Large Language Models: A Plain-Language Overview for Engineers (arxiv.org)

Recursive CLIP: Vision-Language Distillation with Looped Transformer Blocks (arxiv.org)

LiveOption: Evaluating LLM Agents in Structured Option Trading with Nonlinear Payoffs (arxiv.org)

GRASP: Group-Adaptive Hybrid RL-SFT Post-Training for Medical VQA without Reasoning Annotations (arxiv.org)

LLM WikiLeaks: Uncovering URL-Based Capability Extension and Security Circumvention in Agentic Message Boards (arxiv.org)

SentryBench: A Cost-Sensitive Benchmark for Adversarial Robustness of Tool-Using LLM Agents in Public-Safety Deployments (arxiv.org)

Acute Stroke Classification Using Asymmetric Data Standardization and Orthogonal-Focal Loss Optimized CNN-ViT Cross-Attention Fusion: A Multicenter External Validation Study (arxiv.org)

Conviction Without Evidence: Verdict Stability Is Independent of Visual Grounding in Medical Vision--Language Models (arxiv.org)

Multimodal Requirements Traceability Link Recovery in Industrial Settings (arxiv.org)

Efficient Proof Generation for Lean Statements with Multiple Candidate Answers (arxiv.org)

EvolveMAS: A Unified Self-Evolution Framework for Decentralized Multi-Agent Systems (arxiv.org)

CyberDual: AI Attackers versus AI Code Reviewers in Real Software Maintenance (arxiv.org)

TRIAGE: Efficient Vision-Language-Action Inference Needs Both Reuse and Pruning (arxiv.org)

MulFCoder: Framework-conditioned Multi-agent for MLLM-based Multi-framework Front-end Code Generation (arxiv.org)

Spatial Audio Generation and Editing with Executable Plans and Self-Distilling Transfusion (arxiv.org)

ProMo3D: Probing Motion Cues in Frozen 3D Foundation Models (arxiv.org)

Dynamic Multi-Step Decoding for Time-Series Foundation Models (arxiv.org)

ParetoRSI-Gym: Can AI Agents Improve Language Models Across Multiple Domains? (arxiv.org)

Outcome-Aware Agents: When Does Causal Structure Help a Forward Model Foresee the Consequences of Actions? (arxiv.org)

AndroidFlux: Evaluating and Reward Modeling Failure Recovery Capabilities in Mobile Use Multimodal Agents (arxiv.org)

Whether or Where? Tail Calibration Signals Finite-Budget Reward Hacking in Best-of-N Selection (arxiv.org)

SEEK: Skill Extraction with Evolving Knowledge against Frontier Science Agents (arxiv.org)

Xing USP: A Cross-Family Semantic Contract for Information Extraction and Intent-Slot Parsing (arxiv.org)

Improving the Chain-of-Thought Compressibility of Pruned Reasoning Models via Long--Short Contrastive Pruning (arxiv.org)

CoLink: Global Compression and Knowledge Linking for Multi-Turn Dialogue Modeling (arxiv.org)

SHIFT-JEV: Certified Abstention and Gain-Gated Continual Adaptation for LLM-Based Decision Models under Distribution Shift (arxiv.org)

Developer Task Delegation to LLMs and the Trust Factor: A Mixed Methods Study (arxiv.org)

Automated, auditable dating of the Babylonian Astronomical Diaries with language models and astronomical tools (arxiv.org)