Aravind Srinivas: belief

Transcript context

…All right. So what about the query stage? So I type in a bunch of BS. I type poorly structured query. What kind of processing can be done to make that usable? Is that an LLM type of problem? I think LLMs really help there. So what LLMs add is even if your initial retrieval doesn’t have a amazing set of documents, like it has really good recall but not as high a precision, LLMs can still find a needle in the haystack and traditional search cannot, because they’re all about precision and recall simultaneously. In Google, even though we call it 10 blue links, you get annoyed if you don’t even have the right link in the first three or four. The eye is so tuned to getting it right. LLMs are fine. You get the right link maybe in the 10th or ninth. You feed it in the model. It can still know that that was more relevant than the first. So that flexibility allows you to rethink where to put your resources in terms of whether you want to keep making the model better or whether you want to make the retrieval stage better. It’s a trade-off. In computer science, it’s all about trade-offs at the end. So one of the things we should say is that the model, this is the pre-trained LLM, is something that you can swap out in Perplexity. So it could be GPT-4o, it could be Claude 3, it can be Llama. Something based on Llama 3.…

Stored transcript either side of the excerpt. The highlighted words are the published quote; the surrounding text is unedited source, never generated.