Hybrid Search
Hybrid search combines dense vector similarity with sparse keyword search (e.g., BM25), fusing results to capture both semantic matches and exact term overlap for better retrieval.
What is Hybrid Search?
Hybrid search combines dense vector similarity with sparse keyword search (e.g., BM25), fusing results to capture both semantic matches and exact term overlap for better retrieval.
Hybrid search combines dense vector similarity with sparse keyword search (e.g., BM25), fusing results to capture both semantic matches and exact term overlap for better retrieval.
Where is it used?
Weaviate hybrid search, Qdrant sparse vectors, and LangChain `EnsembleRetriever` mix BM25 and dense retrievers; production RAG often uses reciprocal rank fusion (RRF) to merge results.
How to build it
Use `langchain.retrievers.EnsembleRetriever(retrievers=[bm25_retriever, dense_retriever], weights=[0.5, 0.5])` and compare recall against dense-only retrieval on a query set.