Unverified60% confidenceFactTime unknown
Vector retrieval excels at capturing semantic similarity (e.g., 'automobile' and 'car'), while BM25 excels at exact keyword matching (e.g., product model numbers, proper nouns)
2
Sources
60%
Confidence
Medium-term (~90 days)
Relevance
7/2/2026
First Seen
Valid until: 9/30/2026
Sources
Related Claims
Unverified混合检索（Hybrid Retrieval）结合语义向量相似度与BM25关键词匹配，能弥补纯语义检索在精确API名称、变量名等场景的不足77% similarUnverified混合检索（Hybrid Search）结合稠密向量语义相似性与BM25稀疏关键词匹配，两路结果经Re-Ranker模型重排后注入LLM上下文69% similarVerified混合检索（Hybrid Search）将向量检索与BM25关键词检索的结果通过RRF（倒数排名融合）等算法合并，兼顾语义理解与精确匹配69% similarUnverified实际生产中通常会结合向量相似度和关键词匹配两种方式进行混合检索67% similarUnverified单纯的向量语义检索对精确术语（如参数名、错误码）容易失准，推荐向量检索加关键词检索（BM25）的混合方案66% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/55487API
curl https://kongchang.com/api/v1/knowledge/claims/55487MCP
get_claim(id=55487)