Why combine dense (vector) and sparse (BM25) retrieval?
mediumAnswer
- Dense embeddings capture semantic similarity but miss exact keyword matches (rare terms, IDs, product codes, brand names, proper nouns).
- BM25 excels on exact / rare tokens.
- Hybrid search combines both — typical: retrieve top-k with each, fuse ranks via Reciprocal Rank Fusion (RRF) or weighted-sum.
- Consistently 5-15% higher recall than dense-only, essential for enterprise search with product codes, part numbers, medical terms.
- Elastic, Weaviate, Qdrant natively support hybrid.
Check yourself — multiple choice
- Dense alone is best
- Dense misses exact / rare tokens; BM25 handles them; RRF fusion → 5-15% higher recall; essential for enterprise search
- BM25 alone is best
- Impossible to combine
Hybrid search: dense + BM25 fused via RRF → best of both worlds, robust to rare tokens.
#retrieval#hybrid-search
Practise LLMs & GenAI
214 interview questions in this topic.