Beyond Reward Hacking: Causal Rewards for Large Language Model Alignment

Re-assign community