// ⌕ RAG & Vector Retrieval
BM25
Classic keyword ranking by term frequency and rarity.
BM25 关键词检索 2009 Beginner core
IN PLAIN WORDS
Terms frequent in a document but rare in the corpus score high, with saturation and length normalization. Simple yet strong; paired with vectors for hybrid search in RAG.
Used in: Elasticsearch, OpenSearch, hybrid search
01Original paper
The Probabilistic Relevance Framework: BM25 and Beyond
Robertson et al. · Foundations and Trends® in Information Retrieval · 2009
DOI ↗ ✓ Verified · Crossref API doi:10.1561/1500000019
02Where it sits
Learn the left column first; the right column builds on this one.
Learn first
- No prerequisites — a good starting point
Then you can learn
- Nothing in the atlas builds on it yet
Related:Dense Passage Retrieval (DPR)Retrieval-Augmented Generation (RAG)
Left: prerequisites · right: builds on this · dashed: related. Full map →
03Learning resources
Official docs, explainers, courses and reference code. All links checked.
Reference code
04Where the course teaches it
On learning paths:AI Application Engineer
For AI / data
- Verification: verified, checked 2026-10-21.
- JSON: /api/learn/algorithms.json (id:
bm25)