-
Couldn't load subscription status.
- Fork 25.6k
Open
Labels
:Search Relevance/VectorsVector searchVector search>enhancementTeam:Search RelevanceMeta label for the Search Relevance team in ElasticsearchMeta label for the Search Relevance team in Elasticsearch
Description
Description
Digging into #129950 its tricky to get absolutely correct.
As a sibling issue to improve centroid vector scoring (which could be done in addition to indexing the centroids), we should consider quantizing centroids to single bit, and reranking with 4 bit quantization.
When the number of centroids gets large, I would expect this to have a pretty significant impact.
Some things to consider:
- Smallest centroid quantization size should be dictated by the smallest in postings list. Meaning if postings list are 2 bit, centroids should be 2 bit
- We will definitely need to oversample and rescore. We can do something simple and do 3x oversampling by default for single bit and something cleverer later.
- The initial bit quantizations should be 1 bit and 4 bit. Though, once we get to 2bit, I think we should just jump to 7bit for higher fidelity.
Metadata
Metadata
Assignees
Labels
:Search Relevance/VectorsVector searchVector search>enhancementTeam:Search RelevanceMeta label for the Search Relevance team in ElasticsearchMeta label for the Search Relevance team in Elasticsearch