Unverified50% confidenceFactTime unknown
The online pipeline in traditional RAG consists of query rewriting, dual-path retrieval (BM25 + vector), context assembly, and answer generation.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
7/2/2026
First Seen
Valid until: 9/30/2026
Sources
Related Claims
UnverifiedA typical RAG pipeline includes document chunking, vector embedding, vector database storage and retrieval, and final answer generation.76% similarUnverifiedRAG系统的核心工程链路包括文档切分策略、Embedding模型选型与向量数据库部署、混合检索（语义检索+BM25关键词检索）以及Reranker重排序74% similarUnverifiedRAG完整链路包括离线阶段（文档解析→文本分块→Embedding向量化→写入向量数据库）和在线阶段（用户提问→查询向量化→向量检索→召回文档重排序→构建Prompt→LLM生成回答）73% similarVerifiedRAG（检索增强生成）技术架构将文档切片、向量化、存入向量数据库，查询时先检索相关片段再交给模型生成回答73% similarUnverifiedRAG（Retrieval-Augmented Generation，检索增强生成）架构在生成文本前先实时抓取目标网站内容，将检索到的最新信息作为上下文依据注入生成过程73% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/60507API
curl https://kongchang.com/api/v1/knowledge/claims/60507MCP
get_claim(id=60507)