E^2GraphRAG: Streamlining Graph-based RAG for High Efficiency and Effectiveness

TL;DR

E^2GraphRAG introduces a streamlined graph-based retrieval-augmented generation framework that significantly enhances efficiency and maintains high effectiveness through adaptive retrieval and optimized indexing.

Contribution

It proposes E^2GraphRAG, a novel framework that improves indexing speed and retrieval efficiency in graph-based RAG systems using large language models and adaptive strategies.

Findings

Up to 10x faster indexing than GraphRAG

100x speedup in retrieval over LightRAG

Maintains competitive QA performance

Abstract

Graph-based RAG methods like GraphRAG have shown promising global understanding of the knowledge base by constructing hierarchical entity graphs. However, they often suffer from inefficiency and rely on manually pre-defined query modes, limiting practical use. In this paper, we propose E^2GraphRAG, a streamlined graph-based RAG framework that improves both Efficiency and Effectiveness. During the indexing stage, E^2GraphRAG constructs a summary tree with large language models and an entity graph with SpaCy based on document chunks. We then construct bidirectional indexes between entities and chunks to capture their many-to-many relationships, enabling fast lookup during both local and global retrieval. For the retrieval stage, we design an adaptive retrieval strategy that leverages the graph structure to retrieve and select between local and global modes. Experiments show that E^2GraphRAG achieves up to 10 times faster indexing than GraphRAG and 100 times speedup over LightRAG in retrieval while maintaining competitive QA performance.