BM25
The lexical half of hybrid search: a ranking function that scores a document by how often a query term appears in it, discounted by how common the term is across the corpus and by document length. It finds exact strings that embedding similarity blurs, a part number, a company name that is also an ordinary word.
What BM25 does not show
It cannot find a paraphrase, which is why retrieval systems pair it with dense retrieval rather than choosing one.
Terms this definition uses
Hybrid search · Corpus · embedding · retrieval · dense retrieval
Definitions that name BM25
Hybrid search · dense retrieval
Related terms in Answer engines and retrieval
How a generated answer gets made, and what your page is competing for inside it.
AEO · GEO · AI visibility score · share of voice · prompt set · brand mention · RAG · grounding · training data · embedding · chunking · context window · token · hallucination · refusal · AI Overview · answer engine · citation · retrieval · hallucinated attribution · Hybrid search · Re-ranking · Passage ranking · Query fan-out · Chunk boundary · cross-encoder · AI Mode · information gain · dense retrieval
Questions about BM25
What is BM25?
The lexical half of hybrid search: a ranking function that scores a document by how often a query term appears in it, discounted by how common the term is across the corpus and by document length. It finds exact strings that embedding similarity blurs, a part number, a company name that is also an ordinary word.
What does BM25 not show or guarantee?
It cannot find a paraphrase, which is why retrieval systems pair it with dense retrieval rather than choosing one.
Which area of the glossary does BM25 belong to?
Answer engines and retrieval: How a generated answer gets made, and what your page is competing for inside it.
← Chunk boundary · cross-encoder →
See it in the full glossary · 772 terms across 19 areas.
Free browser extension
See this on any site you visit, in one click.
CrawlCheck for Chrome grades every site you visit A–F and shows, from your own browser, what search engines and AI assistants are actually served. Free, no account.