HealthNLP_Retrievers at ArchEHR-QA 2026: Cascaded LLM Pipeline for Grounded Clinical Question Answering

2026-04-29

AI Summary

Although patients have access to electronic health records (EHRs), they often struggle to comprehend the specialized clinical content, highlighting the urgent need for reliable and interpretable clinical question-answering systems. This work proposes a four-stage cascaded architecture based on Gemini 2.5 Pro that sequentially performs query refinement, evidence retrieval, evidence-constrained answer generation, and high-precision many-to-many alignment between answers and supporting evidence. By integrating few-shot prompting for query rewriting, heuristic evidence scoring, and a rigorous alignment mechanism, the approach significantly enhances both interpretability and reliability while maintaining clinical accuracy. Evaluated on the ArchEHR-QA 2026 benchmark, the method achieves top performance in question understanding and demonstrates competitive overall results.

Abstract

Patient portals now give individuals direct access to their electronic health records (EHRs), yet access alone does not ensure patients understand or act on the complex clinical information contained in these records. The ArchEHR-QA 2026 shared task addresses this challenge by focusing on grounded question answering over EHRs, and this paper presents the system developed by the HealthNLP_Retrievers team for this task. The proposed approach uses a multi-stage cascaded pipeline powered by the Gemini 2.5 Pro large language model to interpret patient-authored questions and retrieve relevant evidence from lengthy clinical notes. Our architecture comprises four integrated modules: (1) a few-shot query reformulation unit which summarizes verbose patient queries; (2) a heuristic-based evidence scorer which ranks clinical sentences to prioritize recall; (3) a grounded response generator which synthesizes professional-caliber answers restricted strictly to identified evidence; and (4) a high-precision many-to-many alignment framework which links generated answers to supporting clinical sentences. This cascaded approach achieved competitive results. Across the individual tracks, the system ranked 1st in question interpretation, 5th in answer generation, 7th in evidence identification, and 9th in answer-evidence alignment. These results show that integrating large language models within a structured multi-stage pipeline improves grounding, precision, and the professional quality of patient-oriented health communication. To support reproducibility, our source code is publicly available in our GitHub repository

Problem

• clinical question answering
• electronic health records
• patient understanding
• grounded QA
• health communication

Innovation

• cascaded LLM pipeline
• grounded clinical QA
• evidence retrieval
• query reformulation
• answer-evidence alignment