G-Retriever: Retrieval-Augmented Generation for Textual Graph Understanding and Question Answering
By Xiaoxin He, Yijun Tian, Yifei Sun, Nitesh V. Chawla, Thomas Laurent, Yann LeCun, Xavier Bresson, Bryan Hooi. Neural Information Processing Systems, 2024
316 citations recorded by the source corpus.
Given a graph with textual attributes, we enable users to `chat with their graph': that is, to ask questions about the graph using a conversational interface. In response to a user's questions, our method provides textual replies and highlights the relevant parts of the graph. While existing works integrate large language models (LLMs) and graph neural networks (GNNs) in various ways, they mostly focus on either conventional graph tasks (such as node, edge, and graph classification), or on answering simple graph queries on small or synthetic graphs. In contrast, we develop a flexible question-answering framework targeting real-world textual graphs, applicable to multiple applications including scene graph understanding, common sense reasoning, and knowledge graph reasoning. Toward this goal, we first develop a Graph Question Answering (GraphQA) benchmark with data collected from different tasks. Then, we propose our G-Retriever method, introducing the first retrieval-augmented generation (RAG) approach for general textual graphs, which can be fine-tuned to enhance graph understanding via soft prompting. To resist hallucination and to allow for textual graphs that greatly exceed the LLM's context window size, G-Retriever performs RAG over a graph by formulating this task as a Prize-Collecting Steiner Tree optimization problem. Empirical evaluations show that our method outperforms baselines on textual graph tasks from multiple domains, scales well with larger graph sizes, and mitigates hallucination.~\footnote{Our codes and datasets are available at: \url{https://github.com/XiaoxinHe/G-Retriever}}
Subfields: Artificial Intelligence
View Paper: https://www.semanticscholar.org/paper/a41d4a3b005c8ec4f821e6ee96672d930ca9596c
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
In Artificial Intelligence
Citations in the corpus
- From Local to Global: A Graph RAG Approach to Query-Focused Summarization (2024, background): “…advanced RAG (Gao et al., 2023) where the index is a knowledge graph (KAPING, Baek et al., 2023), where subsets of the graph structure (G-Retriever, He et al., 2024) or derived graph metrics (Graph-ToolFormer, Zhang, 2023) are the objects of enquiry, where narrative outputs are strongly grounded…”
- Retrieval-Augmented Generation for Large Language Models: A Survey (2023, background): “Lastly, there’s a risk of generation models overly depending on augmented information, potentially resulting in outputs that merely reiterate the retrieved content without providing new value or synthesized information.”
- ReMindRAG: Low-Cost LLM-Guided Knowledge Graph Traversal for Efficient RAG (2025, unknown): “Leveraging explicit semantic relations among entities via graph-based text representation, KG-RAG systems often improve overall RAG performance [11, 9, 10, 15, 48, 12].”
- MG2-RAG: Multi-Granularity Graph for Multimodal Retrieval-Augmented Generation (2026, unknown): “To overcome the structural limitations of flat retrieval, recent work has incorporated graph-based representations into RAG [16,28,53,62].”
- Comparing RAG and GraphRAG for Page-Level Retrieval Question Answering on Math Textbook (2025, unknown): “[13] similarly found that GraphRAG tends to surface overly general entities (e.g., “methodology” or “benchmark”) that are not helpful for narrow-scope question answering, underscoring its mismatch with localized retrieval tasks.”
- Retrieval And Structuring Augmented Generation with Large Language Models (2025, unknown): “Building upon GraphToken, G-Retriever [76] extends this approach by incorporating retrieval-augmented generation for improved textual graph understanding.”
- RAS: Retrieval-And-Structuring for Knowledge-Intensive LLM Generation (2025, unknown): “Based on the findings, G-Retriever (He et al., 2024) constructs subgraphs by retrieving relevant entities and relations from a global KG, using these as structured knowledge input for LLMs. GraphRAG (Edge et al., 2024) employs an alternative strategy, extracting a comprehensive global KG from the co”