BGE-Reranker
A reranking model based on a cross-encoder that performs fine-grained scoring of query-document pairs. Its accuracy exceeds that of vector similarity, and it is used in the reranking stage of two-stage retrieval in RAG systems.
Core Facts
Commonly used Rerank models include Cohere Rerank and BGE-Reranker.
Reranking uses a cross-encoder to score query-document pairs one by one, with accuracy far higher than vector similarity but at higher computational cost, so fine ranking is done only on the small number of candidates after coarse ranking, forming a two-stage architecture of coarse ranking plus fine ranking.
Common open-source reranking options include BGE-Reranker, the Cohere Rerank API, and others.
Timeline (last 90 days)
Reranking uses a cross-encoder to score query-document pairs one by one, with accuracy far higher than vector similarity but at higher computational cost, so fine ranking is done only on the small number of candidates after coarse ranking, forming a two-stage architecture of coarse ranking plus fine ranking.
Common open-source re-ranking models include Cohere Rerank, BGE-Reranker, and the ms-marco series.
Common open-source reranking options include BGE-Reranker, the Cohere Rerank API, and others.
Commonly used open-source reranking models include the BGE-Reranker series and models released by BAAI.
Reranking is typically implemented using cross-encoders, which concatenate the query and candidate documents before feeding them in together; it usually reranks only the Top-50 to Top-100 candidates and finally selects Top-3 to Top-5 to pass to the generative model.
All Facts (8)
Commonly used Rerank models include Cohere Rerank and BGE-Reranker.
80%VerifiedReranking uses a cross-encoder to score query-document pairs one by one, with accuracy far higher than vector similarity but at higher computational cost, so fine ranking is done only on the small number of candidates after coarse ranking, forming a two-stage architecture of coarse ranking plus fine ranking.
65%VerifiedCommon open-source reranking options include BGE-Reranker, the Cohere Rerank API, and others.
65%VerifiedReranking is typically implemented using cross-encoders, which concatenate the query and candidate documents before feeding them in together; it usually reranks only the Top-50 to Top-100 candidates and finally selects Top-3 to Top-5 to pass to the generative model.
65%VerifiedReranking models such as Cohere Rerank and BGE-Reranker use a Cross-Encoder architecture to perform fine-grained relevance scoring between queries and documents, and are a key component in optimizing RAG precision.
65%UnverifiedReranker模型（如Cohere Rerank、BGE-Reranker、bce-reranker）同时考虑Query和Document的交互信息，排序精度更高但计算成本也更大
90%UnverifiedCommon open-source re-ranking models include Cohere Rerank, BGE-Reranker, and the ms-marco series.
50%UnverifiedCommonly used open-source reranking models include the BGE-Reranker series and models released by BAAI.
50%