AI News

On the Runway Cascade of Transformers for Language Modeling (arxiv.org)

$\tilde{O}(\sqrt{T})$ Regret and Polylogarithmic Constraint Violation for COCO (arxiv.org)

Center-of-Mass Interpretation of Evolution Strategies (arxiv.org)

Improving Instruction Hierarchy by Learning from Goal-Goal Conflicts (arxiv.org)

When Fusion Order Matters: Insights on Progressive Multimodal Representation Learning for Cross-modal Retrieval (arxiv.org)

Co-Design Bench: Multimodal Collaborative Coding Evaluation for Agentic Models (arxiv.org)

LongSearchEval: Evaluating Long-Horizon Web Search Capabilities of Large Language Models via a Trajectory-Aware Agent Framework with Failure-Node Localization (arxiv.org)

Trustworthy Recursive Self-Improvement in the Era of Large Models: Foundation, Risks, Evaluation and Open Challenges (arxiv.org)

Same Story, Different Lens: A Multi-Agent System for Cross-Outlet News Framing Analysis (arxiv.org)

MuSID-KD: Multigranularity Knowledge Distillation for Generative Recommendation with Semantic IDs (arxiv.org)

Trainable Boundary-Keeping Where No Formal Answer Exists: A Controlled Knock-In Study of Reward Training in Evidence-Sufficiency Judgment (arxiv.org)

Before Blaming the LLM, Diagnose the Workflow: A Role-Level Difficulty Profile for Translating Natural Language into Ontology-Grounded Statements (arxiv.org)

Self-Supervised Pretraining for PHM: What Transfers, What Does Not, and What Federating It Costs (arxiv.org)

Proof-of-Agentness: Towards Verifying Agents on the Web (arxiv.org)

AgenticCity-Bench: Longitudinal Multimodal Observations of Urban Dynamics with Agentic Reasoning (arxiv.org)

Application of Artificial Intelligence in the Management of Climate Change-Related Health Issues in Africa (arxiv.org)

LLM-Enhanced Heterogeneous Graph Learning with Business Semantics for Financial Fraud Detection (arxiv.org)

RSIV-Harness: A Recursive Self-Improvement Harness Framework for Open-Domain Conversational Text-to-Visualization (arxiv.org)

Power-of-Two Leaky Integrate-and-Fire Neurons for Event-Driven Vision-Language-Action Inference (arxiv.org)

AdaTrendLike: Adaptive Hierarchical Detrending and Trend Likelihood Scoring for Time Series Anomaly Detection (arxiv.org)

NeSySpec: Neuro-Symbolic Technical Intelligence for Auditing Engineering Specifications with an Application to Aircraft (arxiv.org)

From Denoising Dynamics to Endpoint Uncertainty in Diffusion Language Models (arxiv.org)

Faithful Explanations under Missing Modalities in Multimodal Dementia Prognosis (arxiv.org)

Beyond Physical Illness: Can Clinical Narratives Reveal Emerging Mental Health Risks in Chronically Ill Patients? (arxiv.org)

MARC: Mechanism-Aware Reflective Correction for Articulated Object Code Generation (arxiv.org)

Oddin: Latent Visual Reasoning in Unified Multimodal Models via Self-Distillation from Image Generation (arxiv.org)

A Canary in the Neural Mine: Catching an Image Classifier's Silent, Confident Failures (arxiv.org)

ProtoEEG: Structured Prototype Alignment for EEG Visual Decoding (arxiv.org)

PhaseBreak: Proactive Temporal Protection against Unauthorized Image-to-Video Animation (arxiv.org)

Logic Hijacking of RAG: Single-Document Poisoning via Local Optimization (arxiv.org)

Physics-Informed Graph-Based Safe RL with Constraint-Specific Critics for Community Battery Coordination in Distribution Network (arxiv.org)

OmniShowcase: Omnidirectional Subject-Consistent Video Generation from Multi-View Images (arxiv.org)

Adapting LLM-SC to Living Sacred Heritage: Domain-Specific Extraction, Semantic Networks and Human Evaluation on Pashupatinath Temple Reviews (arxiv.org)

TrySee: Modular Perception-Enhanced VLM Grounding for Closed-Loop Robot Task Planning in Specialized Domains (arxiv.org)

Towards a Unified Framework of Optimal Input-Side Dimensional Configuration for Deep CTR Prediction (arxiv.org)

MiRL: Error-Targeted Reinforcement Learning for Correcting Terminal Agent Misbehaviors (arxiv.org)

Quantitative Generalization Bounds for Multi-Layer Transformers: A Few Long Strings can Beat Many Short Ones (arxiv.org)

Audit Where Hacks Pay: Budgeted Strong-Verifier Audits Against Reward Hacking in GRPO (arxiv.org)

Can JEV Support Clinical Trial Criteria Assessment? (arxiv.org)

The Watchtower Imperative: Lifespan-Calibrated Evaluation for Frontier AI (arxiv.org)

EEGaze: Multimodal Attention Prediction in Visual Search via EEG-Based Saliency Refinement (arxiv.org)

Do Similarity Measures Measure Similarity? How User Activity Creates Phantom Correlation in Recommender Systems (arxiv.org)

Learning When to Commit: Schedule Distillation for Fast Diffusion Language Models (arxiv.org)

Pretraining Knows What, Simulation Teaches How: Compositional Generalization in a Sim-Trained Robot Agent (arxiv.org)

Barycentric Guidance: Turning Foundation Image Editors into Continuous Affective Controllers (arxiv.org)

Reva: Real-Time Vulnerability Detection through Contextual Probing of Large Language Models (arxiv.org)

Profiling a Frontier Agent with EBench: From Direct Control to Harness with Policy (arxiv.org)

Constrained Decoding: Harmful or Beneficial? Revisiting Its Impact on LLM Performance (arxiv.org)

Residual-Based Adaptive Refinement in Hybrid Variational-Collocation PINNs for Singular Elliptic PDEs with Mixed Dirichlet-Neumann Boundary Conditions (arxiv.org)

I-HITL: Inalienable Human-in-the-Loop (arxiv.org)