SUBQRAG: sub-question driven dynamic graph rag

2025-10-08

AI Summary

Existing Graph RAG approaches struggle to support deep, structured reasoning in complex multi-hop question answering, often leading to evidence omission and error propagation. To address this, we propose SD-GraphRAG, a subquestion-driven dynamic graph-augmented generation framework. First, the input question is decomposed into a verifiable chain of subquestions. Then, for each subquestion, relevant triples are dynamically retrieved and extracted in real time to incrementally expand a knowledge graph. Concurrently, a traceable, structured graph memory is constructed to explicitly aggregate supporting evidence and track reasoning paths. SD-GraphRAG tightly integrates subquestion guidance, dynamic graph construction, and structured reasoning. Evaluated on three multi-hop benchmarks—HotpotQA, 2WikiMQA, and MuSiQue—it achieves an average 4.2% improvement in exact match score, demonstrating substantial gains in reasoning depth, evidence completeness, and interpretability.

Technology Category

• Machine Learning: Graph-based Machine Learning
• Data Mining & Knowledge Management: Graph Mining, Social Network Analysis & Community
• Knowledge Representation and Reasoning: Computational Complexity of Reasoning

Application Category

• Search and Retrieval-Augmented AI: Retrieval-Augmented Generation (RAG) and multi-modal RAG
• Graph Algorithms and Modeling for the Web: Representation, reconstruction, and subgraph or motif discovery in Web-related graphs
• Semantics and Knowledge: Scalable techniques for the creation, curation, publication, maintenance, and consumption of large, Web-based, structured, reusable, knowledge graphs and ontologies

Abstract

Graph Retrieval-Augmented Generation (Graph RAG) effectively builds a knowledge graph (KG) to connect disparate facts across a large document corpus. However, this broad-view approach often lacks the deep structured reasoning needed for complex multi-hop question answering (QA), leading to incomplete evidence and error accumulation. To address these limitations, we propose SubQRAG, a sub-question-driven framework that enhances reasoning depth. SubQRAG decomposes a complex question into an ordered chain of verifiable sub-questions. For each sub-question, it retrieves relevant triples from the graph. When the existing graph is insufficient, the system dynamically expands it by extracting new triples from source documents in real time. All triples used in the reasoning process are aggregated into a "graph memory," forming a structured and traceable evidence path for final answer generation. Experiments on three multi-hop QA benchmarks demonstrate that SubQRAG achieves consistent and significant improvements, especially in Exact Match scores.

Problem

• Enhances reasoning depth for complex multi-hop question answering
• Decomposes questions into verifiable sub-questions for structured reasoning
• Dynamically expands knowledge graphs when existing evidence is insufficient

Innovation

• Decomposes complex questions into verifiable sub-questions
• Dynamically expands knowledge graph with real-time extraction
• Aggregates reasoning triples into structured graph memory