Mitigating Reward Hacking in Reinforcement Learning from Human Feedback via Advantage Sign Robustness

arXiv:2604.02986 · cs.LG, cs.AI, cs.CL · Submitted 2026-04-03 · Read on arXiv

cs.LG, cs.AI, cs.CL

Submitted: 2026-04-03

Updated: 2026-11-07

Terminology

Sources

Related papers