---
title: Reranking — AI Glossary
description: Reranking is a second-stage retrieval step that reorders an initial set of candidate documents by relevance to the query, usually with a more precise but costlier model. It refines the fast first-pass results so the most relevant passages rise to the top before being passed to a generator.
canonical: https://prompeteer.ai/glossary/reranking
---

# Reranking

Reranking is a second-stage retrieval step that reorders an initial set of candidate documents by relevance to the query, usually with a more precise but costlier model. It refines the fast first-pass results so the most relevant passages rise to the top before being passed to a generator.

## Definition

Retrieval pipelines commonly use a two-stage design. A fast retriever — often embedding-based vector search — returns many candidates cheaply, but its ranking is approximate. A reranker then scores each query-document pair more carefully, frequently with a cross-encoder that reads the query and document together, and reorders them.

Reranking improves the precision of what reaches the model in retrieval-augmented generation, which directly affects answer quality and grounding. The trade-off is added latency and compute, so rerankers are typically applied only to a shortlist rather than the whole corpus.

## Sources

- [Nogueira & Cho (2019), Passage Re-ranking with BERT](https://arxiv.org/abs/1901.04085)
- [Cohere — Rerank documentation](https://docs.cohere.com/docs/reranking)

## Related terms

- [retrieval-augmented-generation](https://prompeteer.ai/glossary/retrieval-augmented-generation)
- [semantic-search](https://prompeteer.ai/glossary/semantic-search)
- [embedding](https://prompeteer.ai/glossary/embedding)
- [vector-database](https://prompeteer.ai/glossary/vector-database)