AdaGuard: Enhancing Safety and Policy Compliance with Reasoning-Enabled LLM-As-A-Judge Guardrails
Research
Research News
New papers, breakthroughs and academic AI research
Verify Less, Evolve More: Training Idea-Level Critics for Verification-Efficient ML Evolving Agents
Research
SPW-Nav: A Streaming Panoramic World Model for Language-Guided Navigation
Research
RACER: Reflective Agent Coupling Query Interpretation and Tool-Based Retrieval for Frame Selection in Long Video Understanding
Research
Route-Verify-Vote: Procedure-Conditioned Self-Consistency for Mixed-Domain Reasoning
Research
Tokka-Bench: Evaluating Tokenizers Across 100 Natural and 20 Programming Languages
Research
Beyond the Sycophancy Score: How Task, Model, and Pressure Shape LLM Yielding
Research
SkillSandbox: Skill Verification via Dynamic Scenario Synthesis
Research
Task-Oriented Key-Layer KV Communication for Efficient Latent Multi-Agent Collaboration
Research
Child ASR Adaptation with Adult Retention: An Empirical Study
Research
CoDR: Training-Free Confidence-Drift Remasking for Diffusion Language Models
Research
How Could AI Eliminate Humanity? A Failure-Mode Analysis of Civilizational Risk
Research