CARE-Med混合检索-验证框架:医疗大语言模型中基于NLI的幻觉抑制
Hybrid retrieval–verification framework for hallucination reduction in medical LLMs
研究团队提出CARE-Med框架,利用密集检索与自然语言推理(NLI)验证器降低医疗大语言模型幻觉风险。在MIMIC-III等基准测试中,该框架将基线模型的NLI不支持响应率从32%降至10%,相对降幅达68.8%。相比传统RAG方法,NLI不支持响应率从21%降至10%,相对降幅为52.4%。
- Article
- Open access
- Published: 26 October 2026
Scientific Reports (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
Clinicians use Large Language Models (LLMs) for clinical decision support, medical question answering, and automated document writing. However, these models can hallucinate and pose safety risks. While Retrieval-Augmented Generation (RAG) improves factuality, it does not always produce outputs consistent with supporting evidence. To address these issues, this study reframes NLI-based unsupported-response rate mitigation for generative models as a confidence-aware inference problem and proposes a unified CARE-Med framework for retrieval, generation, and verification. The framework ensures the reliability of its outputs by leveraging dense retrieval, natural language inference verifiers, and a confidence scoring mechanism. The authors validated the anti-NLI-based unsupported-response optimisation method for biomedical large models proposed in this paper against the MIMIC-III, PubMedQA, and BioASQ benchmarks. The method reduces NLI-based unsupported-response rate decreases from 32% for the Baseline LLM to 10% for CARE-Med, corresponding to a 22-percentage-point or 68.8% relative reduction. Compared with conventional RAG, NLI-based unsupported-response rate decreases from 21% to 10%, corresponding to an 11-percentage-point or 52.4% relative reduction. The authors leverage explicit verification and uncertainty reasoning to meet requirements for safe clinical deployment.
Funding
The authors extend the appreciation to the Deanship of Postgraduate Studies and Scientific Research at Majmaah University for funding this research work through the project number R-2026-486.
Author information
Authors and Affiliations
School of Engineering, Anurag University, Hyderabad, 500088, India
Sudheer Reddy
Department of Computer Science, College of Computer and Information Sciences, Majmaah University, Al Majmaah, 11952, Saudi Arabia
Shirina Samreen
Authors
- Sudheer Reddy
- Shirina Samreen
Corresponding author
Correspondence to Shirina Samreen.
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
Reddy, S., Samreen, S. Hybrid retrieval–verification framework for hallucination reduction in medical LLMs. Sci Rep (2026). https://doi.org/10.1038/s41598-026-75909-y
Received: 30 July 2026
Accepted: 09 October 2026
Published: 26 October 2026
DOI: https://doi.org/10.1038/s41598-026-75909-y
Keywords
来源:Nature 健康护理主题 · nature.com