SUBQRAG: Sub-Question Driven Dynamic Graph RAG
This addresses reasoning depth limitations in Graph RAG for complex QA, though it appears incremental as an enhancement to existing Graph RAG approaches.
The paper tackles the problem of incomplete evidence and error accumulation in complex multi-hop question answering by proposing SubQRAG, a sub-question-driven framework that dynamically expands knowledge graphs during reasoning. Experiments on three benchmarks show it achieves consistent and significant improvements in Exact Match scores.
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.