AI News

Probe-and-Resume: When and How to Evaluate Agent Execution (arxiv.org)

Sign Betting for Misspecified Regression: Variance Adaptivity and Optimal Approximation (arxiv.org)

Wrong Votes, Right Gradients: Why Majority-Vote Rewards Track Verifiable RL (arxiv.org)

Do Emergent Misalignment Models Share a Circuit? (arxiv.org)

CogDrift-4D: Hierarchical Multimodal Modeling of Temporal Behavioral Decay for MCI Detection (arxiv.org)

Clinically Interpretable Biological Age Estimation Using CatBoost and SHAP with Continuous and Categorical Features from NHANES (arxiv.org)

SHRED: Self-Play for Heterogeneous Road-Users without Expert Demonstrations (arxiv.org)

Does Compression Imply Generalization? A Minimum Description Length Perspective (arxiv.org)

Memorilla: Latent Semantic Memory for LLMs (arxiv.org)

Same Target, Different Origins: Aligning Latent Rollout for Consistent PDE Emulation (arxiv.org)

Honesty is the Best Policy ... or is it? Demand Management for Probabilistic Service Options (arxiv.org)

Sci$\Pi$vot: Benchmarking Principal-Investigator Judgment at Pivotal Moments of Scientific AutoResearch Agents (arxiv.org)

OmicsBench: a long-horizon multi-turn benchmark of AI agents on research-study-scale bioinformatics (arxiv.org)

Athanor: An Agent-Native System for Autonomous Post-Training Research (arxiv.org)

Learning When to Withdraw: Curriculum Coverage and Rule-Specific Failures in Fine-Tuned Language Agents (arxiv.org)

Recursive Reward Improvement via Adversarial Meta-Prompt Evolution (arxiv.org)

Deep Reinforcement Learning for HPC Job Scheduling: A Statistical Evaluation (arxiv.org)

Long-Context Drafting Does Not Require Long-Context Drafters (arxiv.org)

Denoising Score Matching in Diffusion Models: The Effect of Rank Constraints on Generalization (arxiv.org)

Amortizing Sequential Experimental Design in Reasoning LLMs (arxiv.org)

Task-Phase-Conditioned Expert Routing Is Not a Reliable Proxy for Quantization Sensitivity in MoE FFN Tensors (arxiv.org)

O-RAN-Native Jamming Detection Under Network Heterogeneity (arxiv.org)

Gibbs-Gap Policy Optimization for Off-Policy RL with Verifiable Rewards (arxiv.org)

Optimal Explicit and Implicit Regularization Strengths in High-Dimensional Continual Linear Regression (arxiv.org)

EEG-Based Prediction of Music Enjoyment Using Deep Learning with Explainable Temporal Analysis (arxiv.org)

RDL-Bench: A Benchmark for Task-Aware Question Answering over Relational Databases (arxiv.org)

Evaluating Model Capabilities After Benchmark Saturation (arxiv.org)

Sliced Wasserstein Sketches with Error Certificates (arxiv.org)

Selection-Safe Conformal Early Exit: Distribution-Free Coverage under Adaptive Depth (arxiv.org)

Semantic Alignment of EEG and Text: A Contrastive Learning Framework for Decoding Imagined Speech (arxiv.org)

Optimal Hydra: Model Heterogeneous Federated Learning for LLMs using Data-Informed Hydra-Nets (arxiv.org)

Interpreting Language Model Parameters (arxiv.org)

RoLaQ: Joint Rotation and Range Calibration for 2-Bit KV Caches in Multi-Head Latent Attention (arxiv.org)

ReasonUni: Reasoning-Augmented Reconstruction Advances Unified Multimodal Models (arxiv.org)

Verdict: Encode Once, Decide Many (arxiv.org)

Can LLMs Anticipate Their Cruxes? (arxiv.org)

Linear Independence of Polynomial Compositions with Bounded Input Degree (arxiv.org)

Feasibility Ceilings Before Learning: Sizing and Timing Bounds for Drone-Mesh Interception (arxiv.org)

Privacy-Preserving Text Inference for Health-Misinformation Classification and Semantic Retrieval (arxiv.org)

Purify Before You Align: Improving Representation Alignment for Diffusion Models (arxiv.org)

Global-in-time Spectral Duhamel-Picard Method for Post-Exposure Bake Simulation and Physics-Informed Neural Operators for EUV Lithography (arxiv.org)

AbridgeKV: Compressing Long Reasoning into the KV Cache of Learned Pause Tokens (arxiv.org)

Audio-Visual Dog Identification and Tracking in Real-World Videos (arxiv.org)

D-DOIT: Training-free Adaptation of Discrete Diffusion via Doob's h-Transform (arxiv.org)

CIViC: Evidence Gates for Audio-Conditioned Visual Prediction on Proxy Tasks (arxiv.org)

INFUSE: A Unified Surface Benchmark for Indirect Prompt Injection in Agentic Systems (arxiv.org)

Reference-Guided Turn-Taking Style Transfer for Full-Duplex Spoken Language Models (arxiv.org)

WorldCast: Exploring Intelligent Visual Generation in World State Prediction (arxiv.org)

Open Benchmark of AI Impact on Humans (ImpactBench): an expert-guided, open platform for the holistic evaluation of AI impact on humans (arxiv.org)

ISW-Agent: Integrating LLM-Driven Dialogue with n8n Workflow for Internal Solitary Wave Remote Sensing Image Processing (arxiv.org)