← Back to Glossary
Sparse Retrieval
Traditional keyword search (like BM25 or TF-IDF) that matches exact word occurrences and frequencies between query and documents.
Why it exists
Vector search can miss exact part numbers, product IDs, rare names, or technical jargon. Sparse search excels at exact matches.
How it works
Uses inverted index tables mapping words to document locations, scoring relevance by term frequency and document length.
Related terms
Where you'll use this
roadmap