LlamaIndex Webinar: Tree-Structured Indexing and Retrieval with RAPTOR

RAPTOR is a recent paper that introduces a new tree-structured technique, which hierarchically clusters/summarizes chunks into a tree structure containing both high-level and low-level pieces. During query-time, there is the option of traversing the top of the tree or querying everything as a flat list.

​This resolves issues with naive top-k RAG: naive RAG retrieves low-level details best suited at answering questions over specific facts in the document. But it struggles with any questions over higher-level context.

​In this webinar …

everything indexing list llamaindex low paper query rag retrieval tree webinar