AI News

How language model agents reshape price competition (arxiv.org)

Property Estimation in Geotechnical Databases Using Labeled Random Finite Sets (arxiv.org)

Waiting Buys Exactly Two Moments: Endogenous Identification in Prediction-Driven Allocation (arxiv.org)

CocoRNA: Collective RNA Design with Cooperative Multi-agent Reinforcement Learning (arxiv.org)

Learning Personalized Memory Gating from Delayed One-Sided Feedback (arxiv.org)

Language Models Learn to Improve Language Models (arxiv.org)

MobileInject: Evaluating Reasoning-to-Action Security of MLLM-Based Mobile Agents under Cross-Modal Consistency Injection (arxiv.org)

Designing for Imperfect Prompts: A Simulation-Based Evaluation of an LLM-Based Enterprise Data Assistant Using Knowledge Graphs (arxiv.org)

Explainable Machine Learning Analysis of Temperature Effects on Mechanical Strength of Fiber-Reinforced High-Performance Concrete (arxiv.org)

Logic-Guided Graph Denoising Diffusion Models (arxiv.org)

Universal Semantic Enhancement for Weakly Supervised Anomaly Detection (arxiv.org)

LagNet: Lag-Structured Operators for Long-Term Time Series Forecasting (arxiv.org)

Escaping Parameter Space: Tight Generalization Bounds via Representation Quality (arxiv.org)

Neural Expansion: A Unified Mechanism for How Deep Neural Networks Generalize (arxiv.org)

FlashReFocus: Interactive Image Refocusing in Seconds (arxiv.org)

Preventing Data Hallucinations in Agentic AI for Optical Network Troubleshooting (arxiv.org)

MGE-PO: Entropy-Aware Credit Assignment for Mathematical Reasoning (arxiv.org)

qGAMMA: A Framework for Optimal DNN Mapping on Multi-Precision Accelerators (arxiv.org)

Validating LLM-generated political networks with survey-reported partisan exposure (arxiv.org)

Framing Spatial Hearing within the Free Energy Principle: A Hierarchical Model of Perceptual Learning (arxiv.org)

Blitz: Accelerating the Diffusion Modules of AlphaFold3 and Boltz-2 (arxiv.org)

TriRouter: Sparse-to-Dense Electrode-Space Routing for Robust EEG Motor Imagery Decoding (arxiv.org)

JAXtari: High-Throughput and Easy-to-Modify Arcade Learning Environment (arxiv.org)

Deep LLM layers expose passage-level context in EEG during silent reading (arxiv.org)

Deconstructing Multi-Task Active Learning: The Paradox of Gradient Conflict and Orthogonal Decomposition (arxiv.org)

Explaining Chess and Go in Natural Language (arxiv.org)

Unifying Embodied Foundation Models via a Model-to-Policy Formulation (arxiv.org)

SpikeQ: Spike-Guided Post-Training Quantization for Spiking Neural Networks (arxiv.org)

PersonaCUABench: Benchmarking Computer-Use Agents Serving Diverse Personas (arxiv.org)

Prism Attention: Proposal-Refined Index Sharing Mechanism for Efficient LLMs Inference (arxiv.org)

GRAFT-EEG: Gauge-Aware Physical Field Transport for Cross-Dataset Motor-Imagery EEG (arxiv.org)

ThetreStageAI: Multi-View Retrieval-Augmented Generation for Structured Bengali Theatre Production Planning (arxiv.org)

The Sign Code: The Hidden Binary Nature of Deep Networks (arxiv.org)

How Many, but How Sure? Class Competition as a Confidence Proxy in Multi-Class Density-Based Counting (arxiv.org)

When Magnitude Suffices: Reconstructing Time-Domain-Gated Microwave Signals Without Phase on a Thoracic Phantom (arxiv.org)

Tool Recommendation for Multi-Step Agents: Finding Every Tool a Task Needs (arxiv.org)

NL2SQL with Iterative Self-Refinement (arxiv.org)

Aligning Language Models Towards Counterfactual Invariance (arxiv.org)

BMuon: Budgeting Muon's Spectral Geometry for Structured Compressibility (arxiv.org)

Large Diffusion Models Need More Than EMA (arxiv.org)

Uncertainty-aware AI-enabled framework for predicting hydrogen dispersion in confined spaces to enhance operator situational awareness (arxiv.org)

A Topological Signature of Learning under Noisy Labels (arxiv.org)

Predictable Scaling: Efficient Model Exploration via Scaling-Law Prediction (arxiv.org)

Regularized free deconvolution for spectral repair of resampling distortion in high-dimensional particle filters (arxiv.org)

RECREG: Recovery-based Shielding for Regular Safety Properties in Reinforcement Learning (arxiv.org)

Bridging Compressed Communication and Local Steps via Spectral Method in Federated Learning (arxiv.org)

Residual Stochastic Concept Bottleneck Models (arxiv.org)

Flash RNN-T: Memory-Efficient Joint Network and Transducer Loss Computation (arxiv.org)

Information Estimation with Discrete Diffusion (arxiv.org)

Three LiDARs, Twenty-Five Datasets: Cross-Source Generalization in Tree-Species Recognition (arxiv.org)