Neural Search Talks — Zeta Alpha cover image

ColBERT + ColBERTv2: late interaction at a reasonable inference cost

Neural Search Talks — Zeta Alpha

00:00

How to Find Tokens in a Document Collection

This basically reduces the search footprint of your index, right? Like the average that is. And then like the size of each cluster. It's quite a few clusters, but I don't want to say the wrong number when they find it again. So they say the total number of tokens, so across documents is something like $600 million. They clustered these into about 262,000 clusters, so 2 to the 18th.

Transcript
Play full episode

Remember Everything You Learn from Podcasts

Save insights instantly, chat with episodes, and build lasting knowledge - all powered by AI.
App store bannerPlay store banner