Training Language Models to Follow Instructions with Human Feedback

by OpenAI · free · Last verified 2026-03-17

Presents InstructGPT, which uses Reinforcement Learning from Human Feedback (RLHF) to align GPT-3 with human intent. By fine-tuning on human demonstrations and training a reward model on human preference comparisons, InstructGPT produces outputs that human evaluators prefer to GPT-3 outputs despite having 100× fewer parameters.