Unverified50% confidenceSolutionExact time
父子文档检索（small-to-big）用较小的块提高检索精度，在送入LLM时回溯到更大的父级块以提供充分上下文，实现检索粒度与生成粒度的解耦
1
Sources
50%
Confidence
Long-term
Relevance
8/28/2026
First Seen
Sources
Related Entities
Related Claims
Unverified重排阶段通常对10-50条候选文档进行精排，能够捕捉双编码器遗漏的细微语义关联从而提升检索相关性67% similarUnverified检索与注入层常用技术包括MMR算法、Hybrid Search混合召回、Cross-Encoder重排序及时间衰减加权65% similarUnverifiedVector retrieval relies on vector databases for fuzzy matching and cannot precisely express structural relationships, whereas graph retrieval uses graph databases as the storage medium and is specifically optimized for multi-hop traversal queries.64% similarUnverifiedThe HNSW algorithm achieves hierarchical search by constructing a multi-layer graph structure, approximately reducing the exponential complexity of exact nearest-neighbor search to logarithmic complexity, making it the most mainstream index structure in current vector databases.63% similarUnverified大规模代码库的精准检索融合基于AST的结构化分析、向量嵌入技术和BM25等传统信息检索算法62% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/813712API
curl https://kongchang.com/api/v1/knowledge/claims/813712MCP
get_claim(id=813712)