AI News

SkillRL: Evolving Agents via Recursive Skill-Augmented Reinforcement Learning (arxiv.org)

Hardware-aware Calibrated Clustered Attention for Efficient Visual Geometric Transformers (arxiv.org)

Beyond Nominal Equilibria: Risk-Averse Multi-Population Mean-Field Games (arxiv.org)

The Winner's Curse in LLM Self-Improvement Loops: Selection Noise, Lock-in, and Acceptance Rules (arxiv.org)

Correspondences as Decisions: JevNexus for Decision-Centric Schema Matching (arxiv.org)

FLoRa: Flight-Assisted Data Collection from Duty Cycling LoRa Nodes under Energy Constraints (arxiv.org)

CM-DPO: Constraint-Margin Direct Preference Optimization for LLM Planning (arxiv.org)

Few Bits, One Law: Toward W2A4KV2 (arxiv.org)

FreeEvolve: Learning to Evolve Beyond Fixed Loops (arxiv.org)

Sampling SU(N) gauge theory on a 2D lattice from independent plaquettes via holonomies and corner reweighting (arxiv.org)

Quantize by Drift: Label-Free Mixed-Precision Post-Training Quantization for Text Embedders (arxiv.org)

CADFather: Autonomous CAD Reconstruction through Coordinated Tool Use (arxiv.org)

Workhorse: Learning Robust Whole-Body Humanoid Loco-Manipulation from Human Data (arxiv.org)

From Uncertainty to Action: Learning to Steer LLM Agents (arxiv.org)

Same Text, Different Prediction: Serving-Context Nondeterminism in Text Classifiers (arxiv.org)

On the Computational Complexity of Hidden Markov Model Identification (arxiv.org)

U-Space: Uncovering When and Why Uncertainty Arises in Language Models (arxiv.org)

Towards AI-Generated Music Plagiarism Detection as a Version Identification Problem (arxiv.org)

Talking with Language Models (arxiv.org)

Shared-Roadmap Generation and Evaluator for Multi-Agent Path Planning Using Heterogeneous Graph Neural Network (arxiv.org)

Beyond Explanation: Debugging Medical Imaging Models via Concept Intervention (arxiv.org)

How Fragile Is On-Device Language Model Safety? Localizing Safety-Critical Parameters for Sparse Fault Analysis (arxiv.org)

A method for multimodal analysis of TAIGA experiment data using essential features (arxiv.org)

HULK: Learning Whole-Body Forceful Loco-Manipulation for Humanoids (arxiv.org)

CARE: Certifying Acceleration for Vision-Language-Action Inference (arxiv.org)

Task-Sufficient Contraction: Source Selection for Machine Information Interfaces (arxiv.org)

Learning Unknown Constraints without Unsafe Data via Optimality and Counterfactual Regularization (arxiv.org)

Evaluating Trajectory Features for Routing Final-Layer Attention (arxiv.org)

An Informational Curse of Horizon in Goal-Conditioned Policy Learning (arxiv.org)

Conditional Accuracy Profiles: Diagnosing LLM Judges across Deployment Conditions (arxiv.org)

Consistent Distribution Matching for Data-Free Diffusion Distillation (arxiv.org)

Quad-State Safety Evaluation of Open-Weight Large Language Models on Non-Canonical Inputs (arxiv.org)

CurveTQ: Rotation-Free Trellis Quantization of LLM Weights via Curvature-Weighted Search (arxiv.org)

Patient, Place, Prior (P$^3$): What Counts as Personalization in Medical World Models? (arxiv.org)

The Dichotomy Between Pattern Recognition and Step-by-Step Reasoning (arxiv.org)

Directional Evidence Guided Search-Space Reduction for Exact DAG Learning (arxiv.org)

Q-PACE: Dynamic Precision Allocation for Quantization-Aware Training (arxiv.org)

LayerRoPE: Dynamic Depth-wise Magnitude & Angular Superposition (arxiv.org)

Context-aware Attention-based Gaussian Mixture Models for Vehicular Trajectory Prediction (arxiv.org)

Symmetry-Informed Causal Partial Identification (arxiv.org)

Noise Your Prompt: Noising Conditioning Tokens in Continuous Diffusion Language Models (arxiv.org)

A Cognitive-Aware QML-CRL Framework for Detecting Affinity and Romance-Investment Fraud (arxiv.org)

Reinforcement Learning-Based Traffic Signal Control for IoT-Enabled Intersections (arxiv.org)

A Positive Case for Faithfulness: LLM Self-Explanations Help Predict Model Behavior (arxiv.org)

Machine learning modularity (arxiv.org)

Persistence Paradox in Dynamic Science: Evidence from the Deep Learning Revolution (arxiv.org)

APE: Selective Fine-tuning with Acceptance Criteria for Language Model Adaptation (arxiv.org)

Graph-Based Floor Separation Using Node Embeddings and Clustering of WiFi Trajectories (arxiv.org)

Requirement-Based Testing: Enhancing Reinforcement Learning with Game Theory (arxiv.org)

The Signed Geometry of One-Shot Recourse: On-Path Validity and the Signed-Curvature Criterion (arxiv.org)