Unverified50% confidenceCautionExact time
Cross-encoders are only suitable for re-ranking an already narrowed candidate set (typically 20–50 items), not for scanning the full index.
1
Sources
50%
Confidence
Long-term
Relevance
10/11/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedReranking 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.74% similarUnverified重排阶段通常对10-50条候选文档进行精排，能够捕捉双编码器遗漏的细微语义关联从而提升检索相关性73% similarUnverifiedThe computational complexity of cross-encoders is O(n²); document representations cannot be precomputed and scoring must be done online pair by pair, so cross-encoders are typically only used to rerank the top-k results returned by an initial retrieval stage.65% similarVerified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65% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/998517API
curl https://kongchang.com/api/v1/knowledge/claims/998517MCP
get_claim(id=998517)