False Frontiers: Diagnosing and Mitigating Co-Cheating in Self-Evolving Search Agents Paper • 2609.39102 • Published Sep 30 • 150
The Teacher Is a Direction, Not a Destination: Extrapolating RL-Induced Representation Residuals in On-Policy Distillation Paper • 2609.36484 • Published Sep 29 • 168
MaLiang-Harness: A Programmable Path to Image and Video Generation Paper • 2609.34309 • Published Sep 28 • 89
Terminal-Universe: Turning Agent Trajectories into Scalable Terminal Environments Paper • 2609.04148 • Published Sep 3 • 79