RLHF

Reinforcement learning from human feedback (RLHF) is a machine learning (ML) technique that uses human feedback to improve ML models.