RLHF

Reinforcement Learning from Human Feedback — the technique that made ChatGPT helpful instead of just plausible.

Definition

RLHF (Reinforcement Learning from Human Feedback) is the training process where humans rank AI outputs and the model learns to prefer the higher-ranked ones. It's the difference between a model that produces statistically likely text and one that produces text humans actually find useful. RLHF is the dominant technique behind ChatGPT, Claude, and Gemini.

Examples

  • Comparing two AI responses and picking the better one
  • Writing a "perfect" reference response to a prompt for the model to learn from
  • Flagging unsafe or factually wrong outputs

Related terms

Looking for actual gigs that use RLHF? Browse all live listings or filter by $100+/hr.