Unverified50% confidenceFactExact time
Vector similarity in RAG retrieval is measured using algorithms such as cosine similarity or Euclidean distance.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
7/2/2026
First Seen
Valid until: 9/30/2026
Sources
Related Claims
Unverified向量相似度通常使用余弦相似度（Cosine Similarity）或点积（Dot Product）来衡量，这是 RAG 检索的数学基础80% similarUnverifiedFeature matching computes similarity between extracted vectors and registered features typically using Euclidean distance or cosine similarity78% similarVerified向量相似度计算通常采用余弦相似度，它衡量向量方向的夹角而非绝对距离，对向量长度不敏感70% similarUnverifiedCosine similarity measures the degree of similarity between vectors with a range of [-1, 1], where 1 indicates identical direction69% similarUnverified余弦相似度衡量的是两个向量在高维球面上的角度差异，依赖全局方向信息，忽略嵌入向量各维度之间的局部关系结构64% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/44933API
curl https://kongchang.com/api/v1/knowledge/claims/44933MCP
get_claim(id=44933)