CG-RAG: Research Question Answering by Citation Graph Retrieval-Augmented LLMs

2025-01-25

AI Summary

To address the limitations of existing retrieval-augmented generation (RAG) methods in scientific question answering—specifically, insufficient modeling of complex citation relationships and low retrieval precision that constrain answer quality—this paper proposes a citation-graph-enhanced RAG framework. Our key contributions are threefold: (1) a novel contextualized citation graph representation that explicitly captures both semantic and structural dependencies among scholarly documents; (2) Lexical-Semantic Graph Retrieval (LeSeGR), a unified retrieval method integrating sparse and dense signals with graph neural network encoding; and (3) end-to-end context-aware alignment of graph-structured information into large language model (LLM) generation. Evaluated across multi-domain scientific QA benchmarks, our approach significantly outperforms state-of-the-art RAG baselines, achieving new SOTA results in both retrieval accuracy and answer quality. This demonstrates the critical role of graph-structured retrieval in supporting complex academic reasoning.

Technology Category

• Natural Language Processing: Question Answering
• Data Mining & Knowledge Management: Conversational Systems for Recommendation & Retrieval
• Machine Learning: Graph-based Machine Learning

Application Category

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

Abstract

Research question answering requires accurate retrieval and contextual understanding of scientific literature. However, current Retrieval-Augmented Generation (RAG) methods often struggle to balance complex document relationships with precise information retrieval. In this paper, we introduce Contextualized Graph Retrieval-Augmented Generation (CG-RAG), a novel framework that integrates sparse and dense retrieval signals within graph structures to enhance retrieval efficiency and subsequently improve generation quality for research question answering. First, we propose a contextual graph representation for citation graphs, effectively capturing both explicit and implicit connections within and across documents. Next, we introduce Lexical-Semantic Graph Retrieval (LeSeGR), which seamlessly integrates sparse and dense retrieval signals with graph encoding. It bridges the gap between lexical precision and semantic understanding in citation graph retrieval, demonstrating generalizability to existing graph retrieval and hybrid retrieval methods. Finally, we present a context-aware generation strategy that utilizes the retrieved graph-structured information to generate precise and contextually enriched responses using large language models (LLMs). Extensive experiments on research question answering benchmarks across multiple domains demonstrate that our CG-RAG framework significantly outperforms RAG methods combined with various state-of-the-art retrieval approaches, delivering superior retrieval accuracy and generation quality.

Problem

• Complex Document Relationships
• Information Retrieval
• Answer Accuracy

Innovation

• CG-RAG
• LeSeGR
• Large Language Model