Unverified50% confidenceSolutionExact time
Multi-query retrieval uses an LLM to automatically generate 3–5 semantically equivalent but differently phrased variant queries for the same question, retrieves results for each separately, and takes the union to improve recall.
1
Sources
50%
Confidence
Long-term
Relevance
7/24/2026
First Seen
Sources
上下文工程：让AI Agent不再"胡言乱语"的系统方法
Related Claims
UnverifiedIn 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.74% similarUnverifiedA search engine is an indexing and ranking system that displays multiple sources for users to choose from, whereas an LLM is a synthesis and generation system that integrates and rewrites information before presenting it as a single answer.73% similarUnverifiedAnythingLLM中将聊天模式切换为查询（Query）模式后，模型才会严格只基于检索到的文档片段回答72% similarUnverifiedThe multi-query rewriting strategy involves rewriting the user's original question into 3–5 versions that are semantically equivalent but differently expressed, then concurrently retrieving all paths to expand recall coverage.71% similarUnverifiedLLM 与形式化证明系统结合主要有生成式（Generative）和搜索引导式（Search-Guided）两种范式69% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/607420API
curl https://kongchang.com/api/v1/knowledge/claims/607420MCP
get_claim(id=607420)