#434 – Aravind Srinivas: Perplexity CEO on Future of AI, Search & the Internet

5245.968 - 5271.521 Aravind Srinivas

This way, even if you didn't arrive at the right answer, if you had been given the hint of the right answer, you're trying to like reason what would have gotten me that right answer and then training on that. And mathematically you can prove that it's like related to the variation lower bound with the latent. And I think it's a very interesting way to use natural language explanations as a latent.