Saved in:
| Main Authors: | , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2510.07718 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866918168898830336 |
|---|---|
| author | Li, Jiaoyang Ruan, Junhao Tang, Shengwei Chen, Saihan Chang, Kaiyan Ge, Yuan Xiao, Tong Zhu, Jingbo |
| author_facet | Li, Jiaoyang Ruan, Junhao Tang, Shengwei Chen, Saihan Chang, Kaiyan Ge, Yuan Xiao, Tong Zhu, Jingbo |
| contents | 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. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2510_07718 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | SUBQRAG: Sub-Question Driven Dynamic Graph RAG Li, Jiaoyang Ruan, Junhao Tang, Shengwei Chen, Saihan Chang, Kaiyan Ge, Yuan Xiao, Tong Zhu, Jingbo Computation and Language 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. |
| title | SUBQRAG: Sub-Question Driven Dynamic Graph RAG |
| topic | Computation and Language |
| url | https://arxiv.org/abs/2510.07718 |