Skip to main content
Share Your Experience: Take the 2024 Developer Survey
4 votes

Why do we need RL in RLHF?

Recall that RLHF essentially has three stages: Step 0: Pretrain a text-generating LLM. Step 1: Collect human ratings on text, and from that, learn a reward model $r_{\theta}$ which can provide ...
Enigman's user avatar
  • 173
2 votes

What is the difference betwen fine runing and rlhf for llm?

RLHF is just one possibility of fine-tuning for generative LLMs, which is used to align an LLM to human tastes. However, you could just create a bunch of great data, and fine-tune (take a pretrained ...
Alberto's user avatar
  • 2,248

Only top scored, non community-wiki answers of a minimum length are eligible