AI News

The Entropic Signature of Class Speciation in Diffusion Models (arxiv.org)

Modeling Spectral Energy Shifts in Spatio-Temporal Graph Anomaly Detection (arxiv.org)

REAL: Resolving Knowledge Conflicts in Knowledge-Intensive Visual Question Answering via Reasoning-Pivot Alignment (arxiv.org)

Beyond End-to-End Video Models: An LLM-Based Multi-Agent System for Educational Video Generation (arxiv.org)

FLaG: Fine-Grained Latent Grouping for Hallucination Detection (arxiv.org)

When AI Benchmarks Plateau: A Systematic Study of Benchmark Saturation (arxiv.org)

Semi-Supervised Noise Adaptation: Transferring Knowledge from Noise Domain (arxiv.org)

Generalization of Gibbs and Langevin Monte Carlo Algorithms in the Interpolation Regime (arxiv.org)

Multi-Objective Reference-Aligned Machine Unlearning (arxiv.org)

Statistical Testing on Directed Graphs by Surrogate Data Generation (arxiv.org)

Structure Enables Effective Self-Localization of Errors in LLMs (arxiv.org)

Dynamic Entropy Tuning in Reinforcement Learning Low-Level Quadcopter Control: Stochasticity vs Determinism (arxiv.org)

Boosting RL-Based Visual Reasoning with Selective Adversarial Entropy Intervention (arxiv.org)

FlowTime: Towards Continuous Generative Watch Time Prediction via Flow-based Personalized Priors (arxiv.org)

Efficient Hamiltonian, structure and trace distance learning of Gaussian states (arxiv.org)

Inner Product Aware Quantization: Provably Fast, Accurate, and Adaptive Algorithms (arxiv.org)

ReasonBENCH: Benchmarking the (In)Stability of LLM Reasoning (arxiv.org)

AI-Guided Design and Optimization of Graphite-Based Anodes via Iterative Experimental Feedback (arxiv.org)

Machine Learning Surrogate Modeling for Homogenization of Hyperelastic Materials with Boolean Microstructures (arxiv.org)

Connecting the Dots: Benchmarking Reflective Memory in Long-Horizon Dialogue (arxiv.org)

Distilling Neuro-Symbolic Programs into 3D Multi-modal LLMs (arxiv.org)

Prototypicality Bias Reveals Blindspots in Multimodal Evaluation Metrics (arxiv.org)

A Shared Valence Axis Across Modern LLMs and Human EEG: The Saturation Regularity (arxiv.org)

Safety Alignment of LMs via Non-cooperative Games (arxiv.org)

T-POP: Test-Time Personalization with Online Preference Feedback (arxiv.org)

A Fiber Criterion for Representation Identifiability in Supervised Learning (arxiv.org)

World Models: A Comprehensive Survey of Architectures, Methodologies, Reasoning Paradigms, and Applications (arxiv.org)

AI Sovereignty as National Learning Capacity: A Human-Centered Learning Mechanics Viewpoint on France, the United States, and China (arxiv.org)

Decision-Focused On-Policy Learning for Contextual Linear Optimization with Partial Feedback (arxiv.org)

Verifying Meta-Awareness via Predictive Rewards in Reasoning Models (arxiv.org)

A Mathematical Conflict Framework for Contextual Data Modulation (arxiv.org)

From Reward-Free Representations to Preferences: Rethinking Offline Preference-Based Reinforcement Learning (arxiv.org)

ASKD-Whisper: Adaptive Self-knowledge Distillation for Efficient and Low-Latency Automatic Speech Recognition (arxiv.org)

Deep Learning for Remote Sensing to Improve Flood Inundation Mapping (arxiv.org)

A Pre-Training Analogue of Grokking in Language Models: Tracing Delayed Grammatical Generalization (arxiv.org)

SilentDrift: Exploiting Action Chunking for Stealthy Backdoor Attacks on Vision-Language-Action Models (arxiv.org)

Regularized Offline Policy Optimization with Posterior Hybrid Bayesian Belief (arxiv.org)

The Image Reconstruction Game: Drawing Common Ground Through Iterative Multimodal Dialogue (arxiv.org)

Calibrating Uncertainty for Zero-Shot Adversarial CLIP (arxiv.org)

Uncovering Competency Gaps in Large Language Models and Their Benchmarks (arxiv.org)

InFerActive: Interactive Tree-Based Exploration of LLM Sampling for Safety Evaluation (arxiv.org)

Collaborative Space Object Detection with Multi-Satellite Viewpoints in LEO Constellations (arxiv.org)

SpeedAug: Policy Acceleration via Tempo-Enriched Policy and RL Fine-Tuning (arxiv.org)

RadioMaster: Multi-Agent System for Autonomous Radio Signal Generation (arxiv.org)

Train, Test, Re-evaluate: Schedule-Sensitive Evaluation of Generative Data for Hand Detection (arxiv.org)

On the Difficulty of Learning a Meta-network for Training Data Selection (arxiv.org)

ARCA: Adapter-Residual Credit Assignment When Token Signals Degenerate (arxiv.org)

VLM4VLA: Revisiting Vision-Language-Models in Vision-Language-Action Models (arxiv.org)

CORE-MTL: Rethinking Gradient Balancing via Causal Orthogonal Representations (arxiv.org)

ERICA: Quantifying Replicability of Cluster Analysis (arxiv.org)