What does BM25 stand for and what does it score?

Flashcard·Easy·4.0 · 0·~30s·Asked atDify Linkedin Sambanova·Relevant atElastic Hugging Face Neo4j Qdrant

Attempt it

Best Matching 25. A sparse, term frequency based retrieval scoring function (the 25th iteration in the Okapi BM family). It scores a document by summing per-term contributions weighted by term frequency, inverse document frequency, and document length normalization. It excels at exact-token matching (rare names, IDs, error codes) where dense embeddings smooth over the signal. Hybrid retrieval pairs BM25 with dense embeddings and fuses results via Reciprocal Rank Fusion (RRF).

Your answer

**BM25** stands for Best Matching 25. The 25 is the version number; it's the 25th iteration of the Okapi BM family of ranking functions from the 1990s. Despite being thirty years old, it's still the strongest sparse-retrieval baseline and a core component of modern production search. What BM25 actually computes. For a query and a document, BM25 sums per-term contributions. For each query term, the contribution depends on three factors. - Term frequency (TF): how often the term appears in the document. More occurrences mean a higher score, with diminishing returns. - Inverse document frequency (IDF): how rare the term is across the whole corpus. Rare terms (a specific name, a product code) are worth more than common terms (the, and). - Document length normalization: longer documents are penalized so they don't beat short documents just by accumulating more matches. The data structure is an inverted index: a map from each term to the list of documents containing it. No embeddings, no neural network, no GPU. Just counting and a small amount of math per term. BM25 excels at exact-token matching: rare names, product IDs, error codes, function names, anything where the literal string is the signal. Dense embeddings smooth over these because they map related concepts to nearby vectors, which can dilute the exact-token signal. The modern pattern: hybrid retrieval. Run BM25 and a dense embedder side by side, then fuse the two ranked lists with Reciprocal Rank Fusion (RRF). The hybrid usually beats either alone because BM25 catches the exact matches dense misses, and dense catches the semantic matches BM25 misses.

**BM25** stands for Best Matching 25, the 25th iteration in the Okapi BM family of sparse retrieval scoring functions developed at City University London in the 1990s. It scores a document against a query by summing per-term contributions over the query's terms. The scoring formula has three factors per term: $$\text{score}(D, q) = \sum_{t \in q} \text{IDF}(t) \cdot \frac{f(t, D) \cdot (k_1 + 1)}{f(t, D) + k_1 \cdot (1 - b + b \cdot \frac{|D|}{\text{avgdl}})}$$ The components. - `IDF(t)` is inverse document frequency, rewarding rare terms. Computed as `log((N - n_t + 0.5) / (n_t + 0.5))` where N is corpus size and `n_t` is the number of documents containing the term. - `f(t, D)` is the raw term frequency in the document. - `k_1` (typically 1.2-2.0) controls how fast term-frequency saturates. - `b` (typically 0.75) controls how much document length normalizes the score; `|D|` is the document length and `avgdl` is the corpus average. The data structure is an inverted index. Implementations like Elasticsearch, OpenSearch, Lucene, Tantivy, and the newer pure-Rust LanceDB-BM25 all use this same scoring family. Where BM25 shines: exact-token matching. Rare entity names, product SKUs, error codes, function identifiers. Anything where the literal string is the signal. These are dense-embedding blind spots because contextual embedders smooth over surface form to capture meaning. A query for 'CVE-2024-3094' finds nothing semantically related; it needs the exact token, and BM25 surfaces it instantly. The production pattern is hybrid retrieval. Run BM25 and a dense embedder in parallel, retrieve top-100 from each, fuse with Reciprocal Rank Fusion (RRF). The RRF score is `\sum_i 1/(k + rank_i)` with `k=60`. The hybrid typically improves recall@k by 5-15% over best single retriever and is the default RAG architecture in 2026.

TL;DR

BM25 (Best Matching 25) is a sparse term-frequency retrieval function that scores documents by TF, IDF, and length normalization; it pairs with dense embeddings in hybrid retrieval.

Memory aid

Easy to grasp

Picture an old-school librarian who scores every book by counting how often your search words appear inside it, downweights words that appear in almost every book (like the, and, of), and accounts for how long each book is so a giant encyclopedia doesn't beat a short article by sheer volume. That counting system is BM25. It's been around since the 1990s and is shockingly hard to beat for exact-match cases like rare names, product codes, and error strings. Modern systems usually run BM25 and a smart semantic search side by side and combine their answers.

Key concepts

Embeddings

Concept explanation~2 min read

Everything you need to truly understand this topic: intuition, mechanics, step by step explanation, code, formulas, and worked example. Click to expand.

BM25 is the algorithm that refuses to retire. It was published in the 1990s, predates the deep-learning era of NLP entirely, and yet remains a core component of every serious production RAG system in 2026. Understanding what it scores, why dense embeddings haven't replaced it, and how hybrid retrieval works is foundational to retrieval engineering.

This deep dive walks through the BM25 scoring formula, the inverted-index data structure, the exact-match failure mode of dense embeddings that BM25 covers, and the Reciprocal Rank Fusion pattern that combines the two into the modern hybrid-retrieval architecture.

The acronym and the family

BM25 stands for Best Matching 25. The 25 is a version number; it's the 25th formula in the Okapi BM family of probabilistic relevance ranking functions developed at City University London in the 1990s by Stephen Robertson, Karen Sparck Jones, and colleagues.

The earlier members of the family (BM1, BM11, BM15) experimented with different normalizations and combinations of term frequency and document length. BM25 is the formulation that combined all the lessons into a single ranking function and has held the throne as the strongest pure-sparse baseline for thirty years.

The theoretical foundation is the probabilistic relevance framework: rank documents by the probability of relevance given the query, using assumptions about term independence and a binary relevance model. The BM25 formula is the practical instantiation that performed best across the TREC evaluations of the 1990s and has held up across every retrieval benchmark since.

The family is still alive. Robertson and others have published BM25F (field-weighted BM25 for documents with structured fields like title, body, tags), BM25+ (with a floor on term-frequency contributions), and BM25L (alternative length normalization). BM25 itself remains the default.

The scoring formula in detail

The inverted-index data structure

Where BM25 outperforms dense embeddings

Hybrid retrieval and Reciprocal Rank Fusion

Situations where this technique stops working.

2–4 min · Everything important, quickly.

Real products, models, and research that use this idea.

Elasticsearch and OpenSearch implement BM25 as the default scoring function for full-text search queries.
Anthropic's contextual retrieval architecture for Claude pairs BM25 with dense embeddings and a contextual chunk prefix to boost both retrievers.

What an interviewer would ask next. Try answering before peeking at the approach.

QWhat's the difference between BM25 and TF-IDF, and why does it matter?

TF-IDF scores a document by raw TF times IDF, which gives unbounded score for very high frequency terms and ignores document length. BM25 adds term-frequency saturation (controlled by k_1) so the marginal value of additional occurrences drops off, and length normalization (controlled by b) so longer documents don't beat shorter ones by sheer mass. Both changes materially improve quality on real corpora.

2 more follow-ups an interviewer would ask next. Sign in to reveal them.

Red flags & common mistakes

The phrases that signal junior thinking. Click to expand.

Most common mistake

Treating BM25 as obsolete because dense embeddings exist. It's still strong at exact-token matching and the workhorse of hybrid retrieval, where BM25 + dense + RRF beats either alone.

60 second bullets to scan on the way to the call.

BM25 acronym expansion (Best Matching 25)
Three scoring factors (TF, IDF, length normalization)

Primary sources. Browse if you want the original framing.