AI News

Movable-Element Multi-Functional RIS-Assisted Hybrid RSMA-NOMA: A Constrained Hybrid Reinforcement Learning Approach (arxiv.org)

Graph-dependent pooling: adaptive granularity in hierarchical graph neural networks (arxiv.org)

Safe by Benchmark, Unsafe in Practice: Lessons from 1331 Child-Related AI Incidents (arxiv.org)

EntailLLM: Verifying Agent-Generated Vulnerability Discovery Paths with Domain Knowledge via Logic Programming (arxiv.org)

Hierarchical Task Network planning in the Goal Lifecycle (arxiv.org)

Curriculum-Guided Acquisition of Hierarchical Task Knowledge from LLM-Generated Experience (arxiv.org)

Do Halting Heads Really Save Compute in Looped Language Models? (arxiv.org)

Position, Role or Label? A Controlled Study of Instruction Conflicts in LLM Agents (arxiv.org)

SVD-Attention: Efficiently Analyzing Attention Blocks in Transformers (arxiv.org)

EEGaze: Multimodal Attention Prediction in Visual Search via EEG-Based Saliency Refinement (arxiv.org)

Diagnosing and Mitigating Evidence Misgrounding in Multimodal Knowledge Editing (arxiv.org)

Topic-Level Fairness in Retrieval-Augmented Generation (arxiv.org)

Whale You Go Green? Immediate and Longitudinal Effects of Embodied Conversational AI Avatars on Sustainable Behavioral Tendencies and Pro-Environmental Attitudes (arxiv.org)

Identically Trained Language Models Break on Different Jailbreaks (arxiv.org)

Correlation-Aware Exploration for Long-Trace Reasoning (arxiv.org)

CFBrainAug: Modularity-Induced Counterfactual Brain Network Augmentation for Interpretable fMRI-based Autism Diagnosis (arxiv.org)

EVOLVE: Uncertainty-Aware Multivariate Time Series Anomaly Prediction with Adaptive Prototype Memories (arxiv.org)

Reinforcement Learning with Multi-Step Transition Predictions: A Bellman Theory (arxiv.org)

MARC: Mechanism-Aware Reflective Correction for Articulated Object Code Generation (arxiv.org)

A Sign Law for Entropy Drift under Sampler Mismatch in RLVR (arxiv.org)

NumTrace: Auditing the Number Provenance of Financial Analyst Agents (arxiv.org)

AIPC: Evidence-Guided Agent Workflows for PyTorch Project Deployment (arxiv.org)

Do Similarity Measures Measure Similarity? How User Activity Creates Phantom Correlation in Recommender Systems (arxiv.org)

The Path to NDE 4.0: A Review of Algorithms, Techniques, and Strategies for Autonomous Inspection (arxiv.org)

Does LeWorldModel Preserve Local Action Geometry? (arxiv.org)

Linear Probes Transfer from Base to Distilled Language Models despite Rotation (arxiv.org)

Logic Hijacking of RAG: Single-Document Poisoning via Local Optimization (arxiv.org)

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

Announced Breaks Separate Conformal Reliability from Frequency Calibration (arxiv.org)

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

Seeing the Options: Typed Decisions over Visual Candidates with Reusable Option Memory (arxiv.org)

AnalogPortBench: Benchmarking LLMs on Analog Circuit Porting (arxiv.org)

Human-Feedback-Guided Multi-Agent Inverse Reinforcement Learning for Interactive Driving Behavior Modeling (arxiv.org)

PACE: Research Proving Agent for Adaptive Conjecture Exploration (arxiv.org)

Meta-CLIP: Metadata-Conditioned Contrastive Learning for Zero-Shot Disease Inference (arxiv.org)

More: Mode-Optimized Reasoning with Self-Adaptive Effort for Self-Evolving Agents (arxiv.org)

Reasoning Patterns for Efficient AI Coding: An Empirical Study (arxiv.org)

AGQ: Action-Guided Ultra-Low-Bit Quantization for Vision-Language-Action Models (arxiv.org)

CEPO: A General Framework for Improving Prompt Optimization in Multi-Agent Systems (arxiv.org)

TAD: Distilling Vision-Language-Action Policies from Teacher Trajectories Alone (arxiv.org)

RAQ-MLA: Role-Aware Quantization for Multi-Head Latent Attention Caches (arxiv.org)

Jointly Modeling Preferences and Imperfect Decision-Making in Large Language Models (arxiv.org)

When Generated Information Returns: Error Persistence in LLM-Mediated Information Ecosystems (arxiv.org)

Physics-Informed Graph-Based Safe RL with Constraint-Specific Critics for Community Battery Coordination in Distribution Network (arxiv.org)

AdaPRA: Process Reward Agents with Adaptive Rubrics for Data Science (arxiv.org)

Evaluating Visual Language Models on Visual Grounding for Visually Impaired People (arxiv.org)

Towards a Unified Framework of Optimal Input-Side Dimensional Configuration for Deep CTR Prediction (arxiv.org)

T2D: Target-to-Draft Model Pruning & Adaptation for Efficient Speculative Decoding (arxiv.org)

OptiCom : A Unified Framework for State-Conditioned Composition in LLM-Driven Optimization (arxiv.org)

Power-of-Two Leaky Integrate-and-Fire Neurons for Event-Driven Vision-Language-Action Inference (arxiv.org)