AI News

Spectral Convergence of Random Feature Method in Multiple Dimensions (arxiv.org)

TabScope: Question-Adaptive Scope Selection for Table Question Answering (arxiv.org)

Exploring the Potential of Contrastive Language-Image Pre-training for Multi-Source Remote Sensing Data (arxiv.org)

Reducing Catastrophic Risk from AI with Systematic Monitoring and Evaluation of Rogue AI Progression (arxiv.org)

SHELF: A Synthetic Harness for Multi-Task Bibliographic Benchmarking (arxiv.org)

Neural Video Compression Based on Deformable Temporal Alignment and Difference-aware Fusion (arxiv.org)

Toward Collective-Centric Evaluation of Preference Inference for Participatory Democracy (arxiv.org)

When Optimization Becomes Manipulation: Defending Generative Search against Malicious Generative Engine Optimization (arxiv.org)

PrivateHub: Contrastive Diffusion Model for Private Sensor-Intensive Environment Data Generation (arxiv.org)

Reflect-SQL: A Self-Reflection Based Framework for Text-to-SQL (arxiv.org)

Structure and Implementation of New Practical English Textbooks Driven by Artificial Intelligence (arxiv.org)

Counterexamples as Feedback for Agent Self-Correction (arxiv.org)

ExecRetrieval: Measuring the Functional-Correctness Gap in Code-Embedding Retrieval (arxiv.org)

X-Translator: A Real-Time Multilingual Speaker-Aware Speech-to-Speech Translation System (arxiv.org)

Anonymization, Not Elimination: Utility-Preserved Speech Anonymization (arxiv.org)

Traceable TTS: Toward Watermark-Free TTS with Strong Traceability (arxiv.org)

Clean Engineering, Unstable Measurement: A Preregistered Reliability Failure of Black-Box LLM Observers on Shared Endpoints (arxiv.org)

A Computationally Feasible Framework for Causal Probabilistic Explanation (arxiv.org)

Rethinking On-Policy Distillation of Large Language Models II: One Training Example (arxiv.org)

A Case Study on Emergent Cheating and Whistleblowing in Autonomous Research Swarms (arxiv.org)

Terminal-Universe: Turning Agent Trajectories into Scalable Terminal Environments (arxiv.org)

Efficient Test-Time Adaptation through Human-AI Interaction (arxiv.org)

The Natural Language Interaction Protocol and Standard for AI Agents (arxiv.org)

Environment Evolution for Terminal Agents (arxiv.org)

Epistemic Warrant for LLM Recommendations: Characterizing the Basis for Reliance When Ground Truth Is Unavailable (arxiv.org)

DRACO: Fine-Grained Credit Assignment with Dynamic Rubrics for Long-Horizon Agent Training (arxiv.org)

IRWOZ 2.0: A Large Language Model-driven Dialogue Dataset for Industrial Robot Conversations (arxiv.org)

Instruction Duplication as an Inference-Time Control Primitive (arxiv.org)

FLY-EVAL++: An Evidence-Driven Evaluation Protocol for Safety-Constrained Flight Prediction with Large Language Models (arxiv.org)

InSituMeasure: Probing Situated Measurement Grounding in Industrial Scenes with Multimodal Large Language Models (arxiv.org)

LLM4CKD: Large Language Models for Early Stage Chronic Kidney Disease Screening (arxiv.org)

Free Pause Tokens (arxiv.org)

On the Interaction Between Model Compression and Test-Time Adaptation (arxiv.org)

Coupled Scaling: A Representational Accessibility Framework for Neural Scaling Laws (arxiv.org)

Population-Calibrated Graph Screening at 835-Million-Address Scale, with Label-Free Transfer to New Chains (arxiv.org)

Privacy Leakage in Federated Learning: Gradient-Based Client Identity Inference and Defenses for Inertial Sensing in Vehicular Edge Networks (arxiv.org)

Learning from Scarce Labels: Multi-View Echocardiography for Ejection Fraction Prediction (arxiv.org)

Spurious Advantage Hidden in GRPO (arxiv.org)

Privacy-Preserving Topology-Guided Safety for LLM-Based Multi-Agent Systems via Federated Graph Learning (arxiv.org)

Physics-Informed Neural Network Surrogate for Oxygen Vacancy Dynamics in epitaxial $\mathrm{SrTiO_3}$ on Si memristors via Dynamic Spectral Optimization (arxiv.org)

SurfSpec: Enhancing Off-Target-Agnostic Specificity by Bounding Pocket-Ligand Geometric Mismatch (arxiv.org)

FrOGS: Discrete Neural Sampler for Independent Alloy Configurations Across Chemical Conditions (arxiv.org)

Sequential Beats Joint: On the Interplay between On-Policy Distillation and RLVR (arxiv.org)

Privacy-Preserving Heterogeneous Multi-LLM Federated Inference for Cognitive Diagnosis (arxiv.org)

LLM-Guided Reinforcement Learning for Adaptive NPC Behavior in Multi-Agent Combat Games (arxiv.org)

Hadronic Mono-Z Dark Matter Sensitivity with Flow Matching on CMS Open Data (arxiv.org)

You Can't Escape Your Own Activations : Evaluation Awareness and Multi-Agent Monitoring (arxiv.org)

Evaluating GNNs for Success Prediction in Artist Collaboration Networks (arxiv.org)

Judging LLM-as-a-Judge: Concerning Rubric Artifacts in LLM-based Automated Text Generation Evaluation (arxiv.org)

BharatGather: A Culturally-Informed Benchmark Dataset for Misinformation and Fake News Detection in Indian Public Events (arxiv.org)