Scoring Higher, Answering Worse: Mitigating Reward Hacking in Rubric-Based RL via Protocol-Level Rubrics
Paper • 2609.38847 • Published • 61
Carlo Fernandez
carlofer
AI & ML interests
drug discovery, transfer learning, autonomous driving
Recent Activity
liked a model 3 days ago
autotrust/JEV-35BOrganizations
None yet
upvoted a paper 1 day ago