Human Preferences

Deep Reinforcement Learning from Human Preferences

Notes

Foundational RLHF paper. Learning reward models from human comparisons.