heemin32 commented on PR #13525: URL: https://github.com/apache/lucene/pull/13525#issuecomment-2902248141
I believe this proposal to add a new field for multi-vector support is facing significant challenges primarily because we aim to support HNSW-based search on it. However, if our goal were limited to enabling MAXSIM scoring between two multi-vectors using only exact search (i.e., without KNN indexing for faster retrieval), many of these challenges could be avoided. This exact search approach would still be valuable for re-scoring documents after an initial KNN query retrieves a small candidate set. If we want to represent multi-vectors as an array of vectors within a single document—rather than relying on a parent-child document structure—it might make sense to integrate this into the existing field type. However, such an approach should clearly demonstrate a performance advantage to justify the change. -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: issues-unsubscr...@lucene.apache.org For queries about this service, please contact Infrastructure at: us...@infra.apache.org --------------------------------------------------------------------- To unsubscribe, e-mail: issues-unsubscr...@lucene.apache.org For additional commands, e-mail: issues-h...@lucene.apache.org
