Unverified50% confidenceFactExact time
纠错式检索（CRAG）是2024年由Yan等人提出的自适应检索框架，引入检索评估器将检索结果分为正确、不正确、模糊三类并触发不同处理
1
Sources
50%
Confidence
Long-term
Relevance
8/27/2026
First Seen
Sources
Related Entities
Related Claims
Unverified文本分块策略、检索召回数量（Top-K）、重排序模型（Reranker）是影响RAG系统最终效果的三个关键变量61% similarVerified检索增强生成（RAG）是主流幻觉抑制技术，在模型生成回答前从外部知识库检索相关文档片段作为上下文注入提示词57% similarUnverifiedChunks that are too large dilute retrieval precision, while chunks that are too small result in incomplete semantics; the chunking strategy directly determines the retrieval quality of a RAG system.57% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/810053API
curl https://kongchang.com/api/v1/knowledge/claims/810053MCP
get_claim(id=810053)