Unverified60% confidenceFactTime unknown
知识库底层依赖RAG（Retrieval-Augmented Generation，检索增强生成）技术，将私有文档切片后通过Embedding模型转化为高维向量存入向量数据库
2
Sources
60%
Confidence
Long-term
Relevance
5/31/2026
First Seen
Sources
AI智能体自动生成测试用例实战：MCP Server与Dify落地指南
bilibili老范测试开发
Related Entities
Related Claims
UnverifiedRetrieval-Augmented Generation (RAG) anchors facts by retrieving relevant documents from external knowledge bases before generation, but has limitations including inconsistent retrieval quality and limited context windows78% similarUnverified向量数据库通过将文本转化为高维向量并建立近似最近邻（ANN）索引，是RAG检索增强生成架构的核心基础设施77% similarUnverifiedRAG索引阶段将文档切片并通过Embedding模型转换为向量，存入向量数据库如Faiss、Pinecone、Weaviate76% similarUnverifiedSocrati底层技术涉及RAG（检索增强生成）架构，将用户上传的文档切片后存入向量数据库，再通过语义检索驱动课程内容生成76% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/14730API
curl https://kongchang.com/api/v1/knowledge/claims/14730MCP
get_claim(id=14730)