SMART-RAG:面向NCCN肿瘤指南大语言模型推理的结构感知多跳检索框架
Towards a structure-aware multi-hop retrieval framework for large language model reasoning over complex oncology guidelines
研究提出面向NCCN肿瘤指南的结构感知多跳检索框架SMART-RAG,并在肿瘤选择题与自由回答基准上系统评估。常规RAG受检索噪声影响,可使最多20.69%原本正确的问题出现错误预测偏移。SMART-RAG将准确率提高14.02%,并将检索干扰降低11.17%。
- Article
- Open access
- Published: 14 October 2026
- Dongni Li1,
- Guogang Xu2,
- Haitao Huang1,
- Yanjun Zhang3,4,
- Hui Wang1,5,
- Lei Wu6,
- Yue Wu1,
- Qingyu Meng3,4 &
- …
- Xian Wang2,4,7,8
npj Precision Oncology (2026) Cite this article
We’re sharing this article early to provide faster access to peer-reviewed, accepted research. It is citable and carries a permanent DOI. This version is subject to further edits and will be replaced automatically by the final Version of Record. All legal disclaimers apply.
Abstract
Large language models (LLMs) augmented with clinical guidelines show promise in medical question answering, yet structural parsing and evidence retrieval over complex guidelines remain challenging. We propose SMART-RAG (Structure-aware Multi-hop Augmented Retrieval Technique for RAG), a structure-aware multi-hop retrieval framework for National Comprehensive Cancer Network (NCCN) guidelines. SMART-RAG integrates automated structural parsing, structured chunking with metadata, structure-aware reranking, and multi-hop retrieval to handle flowcharts, tables, annotations, and cross-references with dynamic cross-page and cross-guideline evidence tracing. We systematically evaluated SMART-RAG on both choice and free-response oncology benchmarks constructed from public and manually curated datasets. Results showed that, conventional RAG was susceptible to retrieval noise, which could lead to erroneous prediction shifts in up to 20.69% of originally correct questions. In contrast, SMART-RAG achieved more stable performance improvements, increasing accuracy by 14.02% while reducing retrieval interference by 11.17%. Ablation studies further confirmed the importance of chunking, reranking, and multi-hop retrieval.
Acknowledgements
This work was supported by the Hainan Provincial Natural Science Foundation of China (Grant No. 826QN0920).
Author information
Authors and Affiliations
National Key Lab of Autonomous Intelligent Unmanned Systems, School of Automation, Beijing Institute of Technology, Beijing, China
Dongni Li, Haitao Huang, Hui Wang & Yue Wu
Health Management Institute, The Second Medical Center & National Clinical Research Center for Geriatric Diseases, Chinese PLA General Hospital, Beijing, China
Guogang Xu & Xian Wang
Department of General Surgery, The First Medical Center, Chinese PLA General Hospital, Beijing, China
Yanjun Zhang & Qingyu Meng
General surgery institution, Chinese PLA General Hospital, Beijing, China
Yanjun Zhang, Qingyu Meng & Xian Wang
College of Information Science & Electronic Technology, Jiamusi University, Jiamusi, China
Hui Wang
The State Key Laboratory of Power Systems, the Department of Electrical Engineering, Tsinghua University, Beijing, China
Lei Wu
Department of Health Care, The Hainan Hospital of Chinese PLA General Hospital, Sanya, China
Xian Wang
Department of HaiNan Health Care, Second Medical Center, Chinese People’s Liberation Army General Hospital, National Clinical Research Center for Geriatric Diseases, Beijing, China
Xian Wang
Authors
- Dongni Li
- Guogang Xu
- Haitao Huang
- Yanjun Zhang
- Hui Wang
- Lei Wu
- Yue Wu
- Qingyu Meng
- Xian Wang
Corresponding author
Correspondence to Xian Wang.
Ethics declarations
Competing interests
The authors declare no competing interests.
Additional information
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/.
About this article
Cite this article
Li, D., Xu, G., Huang, H. et al. Towards a structure-aware multi-hop retrieval framework for large language model reasoning over complex oncology guidelines. npj Precis. Onc. (2026). https://doi.org/10.1038/s41698-026-01749-8
Received: 22 October 2025
Accepted: 28 September 2026
Published: 14 October 2026
DOI: https://doi.org/10.1038/s41698-026-01749-8
来源:Nature 健康护理主题 · nature.com