Data Science at Home cover image

Debunking AGI Hype and Embracing Reality (Ep. 233)

Data Science at Home

00:00

The Importance of Human Feedback in Language Learning

"Large language models have no access to reality right they only rely on the language so they can't verify anything whatsoever," he says. "We try to patch it by by imposing some sort of human feedback and then basically slapping them on the head with the hammer called you know reinforcement learning much like the under-kun refers or even just the cherry on top of the cake" He adds that we would probably try to wait to this human generated data a lot heavier than what it gains let's say from this self-supervised stuff but I don't think that's going to work in the long run."

Transcript
Play full episode

The AI-powered Podcast Player

Save insights by tapping your headphones, chat with episodes, discover the best highlights - and more!
App store bannerPlay store banner
Get the app