# Reinforcement Learning from Human Feedback (RLHF)

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

**Field:** Artificial Intelligence  
**Category:** LLMs  
**Also known as:** RLHF, Human Feedback Training

## 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.

## 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.

## 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.

## Common mistake

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

## Three principles

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

## Related terms

- [Reward Model](https://learn.benibix.com/term/ai/reward-model)
- [Reinforcement Learning](https://learn.benibix.com/term/ai/reinforcement-learning)
- [Instruction Tuning](https://learn.benibix.com/term/ai/instruction-tuning)
- [Sycophancy (AI)](https://learn.benibix.com/term/ai/sycophancy)
- [Human Evaluation](https://learn.benibix.com/term/ai/human-evaluation)
- [Constitutional AI](https://learn.benibix.com/term/ai/constitutional-ai)
- [Direct Preference Optimization (DPO)](https://learn.benibix.com/term/ai/direct-preference-optimization)

## Told apart from

- [DPO vs RLHF](https://learn.benibix.com/compare/dpo-vs-rlhf): People think DPO is just a kind of RLHF, and it is not. Both teach a model from pairs of answers that people ranked. RLHF first trains a reward model to score answers. Then it uses trial and error to push the chat model toward high scores. DPO skips both steps and trains the chat model on the pairs straight away. So it is simpler to run.
- [RLHF vs RLVR](https://learn.benibix.com/compare/rlhf-vs-rlvr): People think RLVR is just a new name for RLHF. It is not. In RLHF, people rank answers and a reward model learns their taste, so it can reward any kind of reply. In RLVR, a program checks the answer against a key or runs tests, so it only works where a right answer exists. RLHF shapes tone and help. RLVR trains math and code.

## Taught in

- [Post-Training and RLHF](https://learn.benibix.com/learn/ai-llms/post-training-and-rlhf): LLMs, 8/15

![A prompt gets two answers. A person picks the better one. A reward model learns from many such picks, and the language model is then trained to earn a higher reward.](https://learn.benibix.com/diagrams/llm-rlhf.svg)

---

Source: https://learn.benibix.com/term/ai/reinforcement-learning-from-human-feedback
Author: Behnia Zarei (a developer in Vienna, Austria, who builds and teaches data, AI and business intelligence at BENIBIX)
Publisher: BENIBIX Learn (https://learn.benibix.com/)
Last updated: 2026-11-18
Licence: CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). Quote freely with attribution.
