Unverified50% confidenceFactExact time
相似度检索通常取Top-K（如Top-3或Top-5）最相似的文本块，相似度计算常用余弦相似度或内积
1
Sources
50%
Confidence
Long-term
Relevance
8/21/2026
First Seen
Sources
Related Claims
UnverifiedTop-K检索是从向量数据库中取相似度最高的前K个片段作为候选内容71% similarUnverifiedCosine similarity has a value range of [-1, 1]; in semantic retrieval, attention is typically focused only on the positively correlated range of 0 to 1, with values closer to 1 indicating greater semantic similarity.70% 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.68% similarUnverifiedIn actual production, hybrid retrieval combining both vector similarity and keyword matching is typically used.67% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/782547API
curl https://kongchang.com/api/v1/knowledge/claims/782547MCP
get_claim(id=782547)