Unverified50% confidenceTradeoffExact time
BM25依赖精确词汇匹配，在处理专有名词、产品编号、错误代码等场景表现优异，但无法捕捉语义相似性
1
Sources
50%
Confidence
Long-term
Relevance
8/29/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedBM25 is a bag-of-words model that focuses only on term frequency without understanding word order or semantics, but is more reliable for exact term matching and does not suffer from the semantic drift problem of vector retrieval.79% similarUnverified简单的关键词匹配容易产生大量误判和漏判，而更精准的分类需要借助AI模型本身来做语义判断69% similarUnverified向量嵌入的模糊匹配特性在代码精确定位场景成为噪音来源，容易将语义相似但功能不同的代码误判为相近68% similarUnverifiedSemantic retrieval is based on mathematical metrics such as cosine similarity rather than simple keyword matching.64% similarUnverifiedBM25引入词频饱和和文档长度归一化两个关键饱和函数64% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/817052API
curl https://kongchang.com/api/v1/knowledge/claims/817052MCP
get_claim(id=817052)