T-Retriever: Tree-based Hierarchical Retrieval Augmented Generation for Textual Graphs

TL;DR

T-Retriever introduces a tree-based hierarchical retrieval framework that enhances graph-based retrieval-augmented generation by preserving structure and semantics, leading to improved performance on graph reasoning tasks.

Contribution

It proposes a novel tree-based encoding method with adaptive compression and semantic-structural optimization, addressing limitations of existing graph RAG approaches.

Findings

Outperforms state-of-the-art RAG methods on multiple benchmarks.

Provides more coherent and contextually relevant responses.

Effectively preserves hierarchical and semantic information.

Abstract

Retrieval-Augmented Generation (RAG) has significantly enhanced Large Language Models' ability to access external knowledge, yet current graph-based RAG approaches face two critical limitations in managing hierarchical information: they impose rigid layer-specific compression quotas that damage local graph structures, and they prioritize topological structure while neglecting semantic content. We introduce T-Retriever, a novel framework that reformulates attributed graph retrieval as tree-based retrieval using a semantic and structure-guided encoding tree. Our approach features two key innovations: (1) Adaptive Compression Encoding, which replaces artificial compression quotas with a global optimization strategy that preserves the graph's natural hierarchical organization, and (2) Semantic-Structural Entropy ($S^2$-Entropy), which jointly optimizes for both structural cohesion and semantic consistency when creating hierarchical partitions. Experiments across diverse graph reasoning benchmarks demonstrate that T-Retriever significantly outperforms state-of-the-art RAG methods, providing more coherent and contextually relevant responses to complex queries.