URGen: Uncertainty-guided Retrieval-augmented Generation for Guideline-Grounded Medical Question Answering

Authors: Mengieong Hoi, Chen Zhao, Chaoyu Lei, Huifang Zhou, Zhedong Zheng

Published in Findings of EMNLP, 2026

Recommended citation: Mengieong Hoi, Chen Zhao, Chaoyu Lei, Huifang Zhou, Zhedong Zheng, "URGen: Uncertainty-guided Retrieval-augmented Generation for Guideline-Grounded Medical Question Answering." Findings of EMNLP, 2026.

Abstract: Large language models (LLMs) achieve strong fluency in open-domain question answering, yet often hallucinate in specialized medical QA and lack reliable uncertainty quantification, posing risks in safety-critical applications. We propose Uncertainty-guided Retrieval-augmented Generation (URGen), a training-free framework for guideline-grounded medical multiple-choice QA, where authoritative clinical documents (e.g., consensus statements, treatment protocols) are available and answer correctness depends on precise evidence retrieval. URGen is a four-stage pipeline that combines adaptive retrieval depth with uncertainty calibration: (1) multi-temperature knowledge retrieval to rank candidate keypoints, (2) zero-shot adaptive retrieval depth (ZARD) that expands context when rankings are ambiguous, and contracts it when rankings are consistent, (3) multi-temperature answer generation with aggregation over candidates, and (4) entropy-based uncertainty estimation over the answer distribution to enable abstention on high-uncertainty queries. Extensive experiments on two expert-annotated ophthalmology benchmarks, Thyroid Eye Disease and Graves' Disease, show that URGen improves accuracy by 9.29% and 6.19% over a naive RAG-based Meditron-7B baseline, while achieving 3.01% and 4.57% expected calibration error (ECE). URGen is LLM-agnostic, consistently improving diverse backbones (\eg, LLaMA-3.1 and Mistral) and providing interpretable uncertainty indicators for safety-critical deployment.

@inproceedings{hoi2026urgen,
author = "Hoi, Mengieong and Zhao, Chen and Lei, Chaoyu and Zhou, Huifang and Zheng, Zhedong",
title = "{URGen}: Uncertainty-guided Retrieval-augmented Generation for Guideline-Grounded Medical Question Answering",
abstract = "Large language models (LLMs) achieve strong fluency in open-domain question answering, yet often hallucinate in specialized medical QA and lack reliable uncertainty quantification, posing risks in safety-critical applications. We propose Uncertainty-guided Retrieval-augmented Generation (URGen), a training-free framework for guideline-grounded medical multiple-choice QA, where authoritative clinical documents (e.g., consensus statements, treatment protocols) are available and answer correctness depends on precise evidence retrieval. URGen is a four-stage pipeline that combines adaptive retrieval depth with uncertainty calibration: (1) multi-temperature knowledge retrieval to rank candidate keypoints, (2) zero-shot adaptive retrieval depth (ZARD) that expands context when rankings are ambiguous, and contracts it when rankings are consistent, (3) multi-temperature answer generation with aggregation over candidates, and (4) entropy-based uncertainty estimation over the answer distribution to enable abstention on high-uncertainty queries. Extensive experiments on two expert-annotated ophthalmology benchmarks, Thyroid Eye Disease and Graves' Disease, show that URGen improves accuracy by 9.29\% and 6.19\% over a naive RAG-based Meditron-7B baseline, while achieving 3.01\% and 4.57\% expected calibration error (ECE). URGen is LLM-agnostic, consistently improving diverse backbones (\eg, LLaMA-3.1 and Mistral) and providing interpretable uncertainty indicators for safety-critical deployment.",
booktitle = "Findings of EMNLP",
year = "2026" }