Skip to content
BENIBIX Learn

Reinforcement Learning from Human Feedback (RLHF)

AI💬 LLMsLanguage Model Training

Also known as RLHF, Human Feedback Training

A model learns which answers people like best. Then it gives more of those.

Think of it like

Think of it like a coach who says which of two tries was better. Over time, you learn what they like.

Example

Workers see two replies to 'How do I fix a flat tyre?' and pick the clearer one, 100,000 times over. The model then learns to write the clearer kind.

Why it matters

It is a big reason chat helpers are kind and useful, not odd or rude.

Definition

Reinforcement learning from human feedback trains a model to give the kind of answers people prefer. First, people compare pairs of answers and pick the better one. A reward model learns from those picks to score any answer. Then the language model is trained to write answers that score high. It makes replies more helpful and polite, and less likely to cause harm.
People pick the better answer. A reward model learns their taste, and the language model learns to please it.

Common mistake

Thinking it teaches facts. It teaches what answers people like, which can also teach the model to please rather than be right.

Key takeaways

  1. 1
    People compare answers. Pick the better of two.
  2. 2
    A reward model learns. To score any answer.
  3. 3
    The model chases reward. Writes what scores high.

Compare

Learn it properly