Unverified50% confidenceSolutionExact time
重排模型会对查询和每个候选文档进行更精细的相关性打分，将最相关的文档排在前列，从而提升检索质量
1
Sources
50%
Confidence
Long-term
Relevance
8/27/2026
First Seen
Sources
Related Entities
Related Claims
Verified重排序（Re-ranking）机制在初步检索后引入交叉编码器（Cross-Encoder）对候选文档重新打分，提升输入大模型的上下文质量75% similarUnverifiedIn the knowledge retrieval stage, the user's question is converted into a query vector via an embedding model, similarity is computed against the text chunk vectors in the knowledge base (typically using cosine similarity), and the top-K segments are returned by ranking.71% similarUnverifiedProfessional vector databases typically provide capabilities such as hybrid search, re-ranking, and automatic sharding, which need to be implemented separately in a standard library approach.71% similarUnverifiedReranking uses a Cross-Encoder model to score candidate documents retrieved in the initial retrieval pairwise, then fine-ranks the Top-K to feed into the LLM.71% similarUnverifiedMulti-query retrieval uses an LLM to automatically generate 3–5 semantically equivalent but differently phrased variant queries for the same question, retrieves results for each separately, and takes the union to improve recall.69% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/807777API
curl https://kongchang.com/api/v1/knowledge/claims/807777MCP
get_claim(id=807777)