reranker
A model that reorders a shortlist of retrieved passages by reading each one together with the query.
In more detail
Rerankers, usually cross-encoders, are far more accurate than the first-stage search but too slow to run over a whole corpus, so they see only the top 20 to 100 candidates. Set any score floor in the reranker's own units, not in cosine similarity.
Related terms
- cross-encoder — A model that reads a query and a passage together and outputs one relevance score, instead of embedding each separately.
- hybrid search — Combining keyword and semantic search results, so exact names and paraphrases are both found.
- retrieval-augmented generation — Answering a question by first retrieving relevant passages from your own documents and adding them to the prompt, so the model answers from them.
Taught in
- Keyword And Hybrid Search — AI Engineering: LLM Apps, RAG & Agents
- Measuring Retrieval — AI Engineering: LLM Apps, RAG & Agents
- How RAG Works And When To Use It — AI Engineering: LLM Apps, RAG & Agents
- Chunking And Ingestion — AI Engineering: LLM Apps, RAG & Agents
- Retrieval That Finds The Right Thing — AI Engineering: LLM Apps, RAG & Agents
References
- Sentence Transformers: Retrieve & Re-Rank
- Hugging Face: cross-encoder/ms-marco-MiniLM-L6-v2 model card