RLHF is Reinforcement Learning from Human Feedback. It usually refers to fine tuning language models using data labelled by humans. Hugging face have a good o

RLHF is Reinforcement Learning from Human Feedback. It usually refers to fine tuning language models using data labelled by humans. Hugging face have a good overview in this article: https://huggingface.co/blog/rlhf https://huggingface.co/blog/rlhf