Unverified50% confidenceFactExact time
Parent-document retrieval is a retrieval pattern targeting the chunk-boundary problem: at indexing time small chunks are split for embedding matching, and upon a hit the larger unit to which they belong (the parent document or parent paragraph) is returned.
1
Sources
50%
Confidence
Long-term
Relevance
10/19/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedParent–child retrieval uses small chunks for precise matching, then locates and returns the larger section to which the matched chunk belongs to the answer model, thereby separating 'finding the location' from 'providing context'.78% similarUnverified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64% similarUnverified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63% similarUnverifiedDegraded queries can be sliced and analyzed along dimensions such as query type, evidence location, and retrieval stage in order to attribute the cause.60% similarUnverifiedThe authors recommend grouping by source document before assigning splits or splitting by grouping by "source family"; the splitting unit should reflect how information flows across boundaries, rather than choosing the unit that makes the table easiest to divide.59% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1065210API
curl https://kongchang.com/api/v1/knowledge/claims/1065210MCP
get_claim(id=1065210)