Scoring Higher, Answering Worse: Mitigating Reward Hacking in Rubric-Based RL via Protocol-Level Rubrics Paper • 2609.38847 • Published 26 days ago • 41
Groupwise Agentic Grading and Advantage Redistribution for Code Agent RL Paper • 2609.32577 • Published about 1 month ago • 54
VisionHOPE: Visual Backbones as Self-Modifying Learning Systems Paper • 2609.33325 • Published 29 days ago • 345