--- title: "What is reranking?" canonical: "https://xerj.org/science/reranking" updated: "2026-11-14" source: "content/science/reranking.md" --- # What is reranking? **Short answer** (updated 2026-11-14): What is reranking? It is a second stage after retrieval. A fast ranker, such as BM25, first returns a short candidate list. A slower model then reorders those candidates, and does not walk the whole index. XERJ's rerank stage once shipped at 0.3822 nDCG@10, then scored 0.8299 after a fix. **Key number:** XERJ's rerank stage once shipped scoring 0.3822 nDCG@10 while plain BM25 scored 0.7750. Fixed, it scored 0.8299. (as of 2026-10-18; source: [/blog/does-xerj-beat-jev](/blog/does-xerj-beat-jev)) **TL;DR** Reranking is a second stage after retrieval. A fast ranker returns a short candidate list. A slower model then reorders those candidates, and does not walk the whole index again. ## Agent prompt ```text Act as a coding agent. Read https://xerj.org/llms.txt, install XERJ, start a local node, index one folder I point you to, and run one search so I can see a cited passage. ``` *Figure: The 0.3822, 0.7750 and 0.8299 nDCG@10 figures are from the published Does XERJ beat Jev? post, also told on the 100-contributors page. This walkthrough does not add a new run.* ## How it works The first stage has to be cheap enough to search a large index. BM25 is a usual choice. It returns a short list, for example the top tens or hundreds of passages. The second stage looks only at that list. It can use a slower model, because the list is small. The output is the same passages in a new order. Nothing new is retrieved. ## How XERJ uses it XERJ can rerank a first-stage list. The published post about that feature is honest about a miss. In the [Does XERJ beat Jev?](/blog/does-xerj-beat-jev) post, the rerank stage shipped scoring 0.3822 nDCG@10 while plain BM25 scored 0.7750. One request shape was wrong. Fixed, it scored 0.8299. The loss stayed in the post. This page does not add a new run. For first-stage retrieval, see [BM25](/science/bm25) and [hybrid search](/science/hybrid-search). ## History and research A widely cited early method is Nogueira and Cho, [Passage Re-ranking with BERT](https://arxiv.org/abs/1901.04085) (arXiv:1901.04085, 2019). That paper shows a pretrained language model can reorder a candidate list for question answering. XERJ's published numbers above are from the product post, not from that paper. ## Guides See [rerank search results with a calibrated judge](/answers/rerank-search-results-calibrated-judge), [Cohere RAG with XERJ](/blog/cohere-rag-with-xerj) and the [Cohere integration](/integrations/cohere). ## Related terms [BM25](/science/bm25), [hybrid search](/science/hybrid-search), and [reciprocal rank fusion](/science/reciprocal-rank-fusion). ## FAQ ### What is reranking? It is a second-stage step that reorders a short candidate list from a first-stage retriever, such as BM25, without walking the whole index again. ### How is this different from hybrid search? Hybrid search merges two first-stage lists. Reranking takes one candidate list and reorders it. You can use both, in that order. ### What numbers has XERJ published? Rerank once shipped at 0.3822 nDCG@10 while plain BM25 scored 0.7750. One request shape was wrong. Fixed, it scored 0.8299. The loss stayed in the post. ### When should I skip reranking? If the first stage already puts the right passage first, a second stage may not be worth the cost. The published rc.80 gates recorded which features passed and which failed. ## Evidence - XERJ's rerank stage once shipped scoring 0.3822 nDCG@10 while plain BM25 scored 0.7750. Fixed, it scored 0.8299. — [/blog/does-xerj-beat-jev](/blog/does-xerj-beat-jev) ## Related - [What is hybrid search?](/science/hybrid-search) - [BM25 vs vector search](/science/bm25) - [Jev measured, wire spoken](/blog/does-xerj-beat-jev) - [How to run hybrid search in XERJ](/answers/what-is-hybrid-search) - [What is agentic memory?](/science/agentic-memory) (same intent: Science) - [What is context engineering?](/science/context-engineering) (same intent: Science) - [What is Corpus Hub?](/science/corpus-hub) (same intent: Science)