Instagram's recommendation algorithm does not have a rich, human-legible semantic understanding of users' interests; it relies on large embedding models producing illegible high-dimensional vectors, and only recently have LLMs made it possible to translate those vectors into human-readable descriptions.

34:48the algorithm than there is. Most of what's really driven the progress in the world of recommenders over the last 5 10 years have been, you know, these large embedding models and these other techniques that basically produce artifacts that cannot be read by people. They're not legible. They're like giant vectors. It's like, sure, I can show you the vector, but it's just going to be a bunch of numbers

35:11in like a seven dimensional space. It's like, and so when when we talk about does the algorithm know something, usually we think in these more semantic terms. It knows I like surfing and it's like it doesn't. It just has this big ass number that happens to correlate with surfing. Um, that said, I think that is starting to change, right? I think that what one of the things that LLMs are enabling

35:40is they can describe in you know words you know English for or whatever language you prefer what some of those previously illeible artifacts um are at least approximate to if not mean directly right so this is like the thing I've been really I posted about this this week this thing called your algorithm basically the idea is we take a look at all of the stuff that you've

36:05interacted with. And then you know we all of that is in an embedding space. You can think of embedding space as a map. You can map a bunch of videos into the same map. And so videos that are close or similar. And now we can just have an LLM just be like describe that part of the map. And it can be like oh that is like deep pourover coffee

36:25snobbery. Um and that's kind of amazing. >> That is so cool >> that like you can ask the LM to look at these numbers and extrapolate. here's like the topic that you're interested in. >> Yeah. Or look at the videos. >> And so the way you um both So you can also embed concepts into that same space. And so I mean embeddings are really the underlying technology

36:46underneath LLMs, right? That's how all the whole thing works. And so you know, so what what we what we what we do now is we let you, you know, quote unquote see your algorithm. You can see what topics we think you're interested in. Um and you can adjust it. You can add and remove things, but the idea here giving people some agency back in a world where, you know, these social media apps