Comparison
Hybrid searchvsLexical search / BM25
Hybrid search
you run keyword search and vector search over the same corpus and merge the two result lists, because each one finds what the other misses.
Running lexical and semantic search together and merging results, commonly with reciprocal rank fusion. The pragmatic production default: semantic recall for paraphrases, lexical precision for exact terms.
Full entry →Lexical search / BM25
someone searched for an exact error code and semantic search returned five conceptually similar pages that did not contain it.
Classic keyword search, with BM25 as its standard ranking function: term frequency weighted against document frequency and length. Beats semantic search on exact identifiers, error codes and names, which is precisely what engineers search for.
Full entry →