ModelRefs / RAPTOR — AI Glossary
RAPTOR — AI Glossary
A recursive document summarization and indexing method building a hierarchical tree of summaries for multi-level retrieval.
Overview
RAPTOR (Sarthi et al. 2024, Stanford) clusters document chunks, summarizes each cluster, then recursively clusters and summarizes summaries, building a tree. Retrieval queries all levels, returning both fine-grained chunks and high-level summaries. Improves performance on multi-hop and global-synthesis QA tasks over flat retrieval.
Reference details
| Topic | rag |
|---|---|
| Last reviewed | 2026-06-24 |
Related terms
Primary source
Continue your research
Use these connected ModelRefs sections to compare alternatives, inspect implementation paths, and review the evidence and governance boundaries relevant to RAPTOR — AI Glossary.
Frequently asked questions
What is RAPTOR?
A recursive document summarization and indexing method building a hierarchical tree of summaries for multi-level retrieval.
What concepts are related to RAPTOR?
Closely related concepts include modular rag, query decomposition, advanced rag.