Hierarchical Abstract Tree for Cross-Document Retrieval-Augmented Generation

TL;DR

The paper introduces $ ext{ extbackslash}Psi$-RAG, a hierarchical tree-based retrieval-augmented generation framework that improves cross-document multi-hop question answering by adaptive indexing and multi-granular retrieval strategies.

Contribution

It proposes a novel hierarchical abstract tree index and a multi-granular retrieval agent to enhance cross-document retrieval in RAG systems, addressing key limitations of existing methods.

Findings

Outperforms RAPTOR by 25.9% in F1 score on cross-document QA benchmarks.

Outperforms HippoRAG 2 by 7.4% in F1 score.

Supports diverse tasks from token-level QA to document summarization.

Abstract

Retrieval-augmented generation (RAG) enhances large language models with external knowledge, and tree-based RAG organizes documents into hierarchical indexes to support queries at multiple granularities. However, existing Tree-RAG methods designed for single-document retrieval face critical challenges in scaling to cross-document multi-hop questions: (1) poor distribution adaptability, where $k$-means clustering introduces noise due to rigid distribution assumptions; (2) structural isolation, as tree indexes lack explicit cross-document connections; and (3) coarse abstraction, which obscures fine-grained details. To address these limitations, we propose $\Psi$-RAG, a tree-RAG framework with two key components. First, a hierarchical abstract tree index built through an iterative "merging and collapse" process that adapts to data distributions without a priori assumption. Second, a multi-granular retrieval agent that intelligently interacts with the knowledge base with reorganized queries and an agent-powered hybrid retriever. $\Psi$-RAG supports diverse tasks from token-level question answering to document-level summarization. On cross-document multi-hop QA benchmarks, it outperforms RAPTOR by 25.9% and HippoRAG 2 by 7.4% in average F1 score. Code is available at https://github.com/Newiz430/Psi-RAG.