The Sentience Institute Podcast cover image

Raphaël Millière on large language models

The Sentience Institute Podcast

00:00

The History of Natural Language Processing With Computers

There were two different approaches to natural language processing with computers. One was heavily inspired by theoretical linguistics and the works of people like Num Chomsky. The other paradigm that was developed in parallel, you could call it the stochastic or statistical approach. And so initially that work was done more in the field of speech recognition. But eventually there was this idea that emerged as a very powerful intuition that people call the distributional hypothesis. That's essentially the intuition that led with a lot of intermediate steps to current language models.

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