NodeRAG: Structuring Graph-based RAG with Heterogeneous Nodes

2025-04-15

AI Summary

Existing graph-augmented RAG methods suffer from coarse-grained homogeneous graph designs and fragmented integration of graph algorithms, resulting in limited expressivity, poor scalability, and performance bottlenecks. This paper proposes a heterogeneous-node-centric graph modeling framework, introducing the first LLM-friendly graph indexing and reasoning paradigm that natively supports multi-granularity semantic modeling and seamless embedding of diverse graph algorithms—including knowledge graph construction, heterogeneous GNNs, and lightweight subgraph retrieval. By synergistically integrating heterogeneous graph neural networks with collaborative prompt engineering, our approach enables efficient multi-hop reasoning while minimizing retrieval token generation. On multi-hop question-answering benchmarks and head-to-head evaluations against GraphRAG and LightRAG, it reduces query/indexing latency by 37%, decreases storage overhead by 42%, and improves accuracy by 11.6%.

Technology Category

• Machine Learning: Large Multimodal Models (LMMs)
• Data Mining & Knowledge Management: Graph Mining, Social Network Analysis & Community
• Knowledge Representation and Reasoning: Computational Complexity of Reasoning

Application Category

• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Search and Retrieval-Augmented AI: Retrieval-Augmented Generation (RAG) and multi-modal RAG

Abstract

Retrieval-augmented generation (RAG) empowers large language models to access external and private corpus, enabling factually consistent responses in specific domains. By exploiting the inherent structure of the corpus, graph-based RAG methods further enrich this process by building a knowledge graph index and leveraging the structural nature of graphs. However, current graph-based RAG approaches seldom prioritize the design of graph structures. Inadequately designed graph not only impede the seamless integration of diverse graph algorithms but also result in workflow inconsistencies and degraded performance. To further unleash the potential of graph for RAG, we propose NodeRAG, a graph-centric framework introducing heterogeneous graph structures that enable the seamless and holistic integration of graph-based methodologies into the RAG workflow. By aligning closely with the capabilities of LLMs, this framework ensures a fully cohesive and efficient end-to-end process. Through extensive experiments, we demonstrate that NodeRAG exhibits performance advantages over previous methods, including GraphRAG and LightRAG, not only in indexing time, query time, and storage efficiency but also in delivering superior question-answering performance on multi-hop benchmarks and open-ended head-to-head evaluations with minimal retrieval tokens. Our GitHub repository could be seen at https://github.com/Terry-Xu-666/NodeRAG.

Problem

• Enhancing graph-based RAG with heterogeneous node design
• Improving integration of graph algorithms in RAG workflows
• Optimizing performance in indexing, querying, and QA tasks

Innovation

• Introduces heterogeneous graph structures for RAG
• Ensures seamless integration of graph-based methodologies
• Improves performance in indexing and querying