HexaShell: A Thin Harness for Exploring the Limits of VLM Agents as Robot Coding Agents Paper • 2610.23479 • Published 15 days ago • 125
view article Article autotrust/JEV-27B-VL: a decision model that learned to see without a single image of training autotrust • Sep 30 • 44
The Teacher Is a Direction, Not a Destination: Extrapolating RL-Induced Representation Residuals in On-Policy Distillation Paper • 2609.36484 • Published Sep 29 • 208
False Frontiers: Diagnosing and Mitigating Co-Cheating in Self-Evolving Search Agents Paper • 2609.39102 • Published Sep 30 • 235
FlowBalance: Verifier-Grounded Self-Improvement from On-Policy Reasoning Experience Paper • 2609.03241 • Published Sep 3 • 47