TalkRL: The Reinforcement Learning Podcast cover image

John Schulman

TalkRL: The Reinforcement Learning Podcast

CHAPTER

WebGPT - Is There a Line Between Knowledge and Understanding?

WebGPT is a new way of training language models to follow instructions with human feedback that was we young at all. Some people are advocating for very small language models that have no external knowledge aside from language, would be the extreme position. And then other people have talked about language models that just know everything as opposed to having an external knowledge source. So I think it is a little hard to separate knowledge, factual knowledge from understanding.

00:00
Transcript
Play full episode

Remember Everything You Learn from Podcasts

Save insights instantly, chat with episodes, and build lasting knowledge - all powered by AI.
App store bannerPlay store banner