logoalt Hacker News

yorwbatoday at 12:16 PM1 replyview on HN

Most of the fine distinctions are already lost when a document is processed through a pile of linear algebra to turn it into a fixed-size list of floating-point numbers, as you can see from the NDCG@10. Vector search is not a tool for fine distinctions. It's a tool for reducing a large pile of documents to a smaller selection of candidates, which you can then check individually with some more expensive method.


Replies

breadislovetoday at 3:44 PM

The ndcg loss is minimal 90.26 -> 89.65. This means it maintains most of the quality.