Graph Retrieval-Augmented Generation: A Survey
By Boci Peng, Yun Zhu, Yongchao Liu, Xiaohe Bo, Haizhou Shi, Chuntao Hong, Yan Zhang, Siliang Tang. ACM Trans. Inf. Syst., 2024
489 citations recorded by the source corpus.
Recently, Retrieval-Augmented Generation (RAG) has achieved remarkable success in addressing the challenges of Large Language Models (LLMs) without necessitating retraining. By referencing an external knowledge base, RAG refines LLM outputs, effectively mitigating issues such as “hallucination,” lack of domain-specific knowledge, and outdated information. However, the complex structure of relationships among different entities in databases presents challenges for RAG systems. In response, GraphRAG leverages structural information across entities to enable more precise and comprehensive retrieval, capturing relational knowledge and facilitating more accurate, context-aware responses. Given the novelty and potential of GraphRAG, a systematic review of current technologies is imperative. This article provides the first comprehensive overview of GraphRAG methodologies. We formalize the GraphRAG workflow, encompassing Graph-Based Indexing, Graph-Guided Retrieval, and Graph-Enhanced Generation. We then outline the core technologies and training methods at each stage. Additionally, we examine downstream tasks, application domains, evaluation methodologies, and industrial use cases of GraphRAG. Finally, we explore future research directions to inspire further inquiries and advance progress in the field. In order to track recent progress, we set up a repository at https://github.com/pengboci/GraphRAG-Survey.
Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition
View Paper: https://www.semanticscholar.org/paper/9ab45aa875b56335303398e84a59a3756cd9d530
Citations in the corpus
- A Survey on the Memory Mechanism of Large Language Model-based Agents (2024, unknown): “GraphRAG approaches [29, 123, 184] leverage extensive LLM-driven operations (e.g., entity recognition and relation extraction) to transform external text corpora into knowledge graphs, allowing abstractive summarization of relevant information through structural clustering.”
- MARS: Multi-hop Adaptive Retrieval and SPARQL Generation for KGQA (2026, unknown): “Graph-based RAG approaches [17,40] often retrieve and reason over graph neighborhoods or raw triples.”
- A Matryoshka Hierarchical RAG for Efficient Multi-Hop Question Answering (2026, unknown): “Graph-based approaches [21] construct a KG offline starting from the entities and relationships extracted from documents.”
- Comparing RAG and GraphRAG for Page-Level Retrieval Question Answering on Math Textbook (2025, unknown): “This discrepancy underscores the importance of evaluating embedding models on domain-specific tasks—such as textbook QA—rather than relying solely on general leaderboard standings.”
- RAG vs. GraphRAG: A Systematic Evaluation and Key Insights (2025, unknown): “Beyond leveraging existing graphs, an emerging line of work extends GraphRAG to text-based tasks by constructing graphs from unstructured documents, with reported benefits for global summarization [5], planning [23], and reasoning [12].”
- ArchRAG: Attributed Community-based Hierarchical Retrieval-Augmented Generation (2025, unknown): “Despite their remarkable success in question-answering (QA) tasks, they may still generate wrong answers due to a lack of domain-specific and real-time updated knowledge outside their pre-training corpus [50].”
- SimGRAG: Leveraging Similar Subgraphs for Knowledge Graphs Driven Retrieval-Augmented Generation (2024, unknown): “The intricate structures of knowledge graphs (KGs) present significant challenges to traditional RAG pipelines, prompting the development of various techniques for graph-based indexing, retrieval, and generation (Peng et al., 2024).”