The Teacher Is a Direction, Not a Destination: Extrapolating RL-Induced Representation Residuals in On-Policy Distillation Paper • 2609.36484 • Published 13 days ago • 553
False Frontiers: Diagnosing and Mitigating Co-Cheating in Self-Evolving Search Agents Paper • 2609.39102 • Published 12 days ago • 603
AgentGrad: Intervention-guided Prompt Optimization for Multi Agent Systems Paper • 2609.08572 • Published Sep 8 • 86