A Novel Approach Based on Integrating Small Language Models and Retrieval-Augmented Generation for Medical Question Answering

Published online: Jul 17, 2026 Full Text: PDF (1.63 MiB) DOI: https://doi.org/10.24138/jcomss-2026-0036
Cite this paper
Authors:
Bao-Gia Nguyen, Quang-Hung Le

Abstract

This paper proposes a novel approach which com bines small language models with retrieval-augmented gener ation in medical question answering to provide accurate and comprehensible information. Our method extracts relevant evi dence from external knowledge and converts them into vector embeddings which are used for high-fidelity semantic retrieval. The small language model subsequently synthesizes the retrieved evidence into fluent, context-aware responses. Extensive experi ments conducted on the PubMed benchmark dataset, the results show competitive performance to larger language models while being far more suitable for deployment in resource-constrained environments. Moreover, the proposed method supports efficient domain knowledge updates without the need for extensive re training. Our implementation is available in the following GitHub repository: https://github.com/LeoBaoNguyen12/RagSLM-MQA.

Keywords

Small Language Models, Retrieval-Augmented Generation, Medical Question Answering, RagSLM-MQA
Creative Commons License 4.0
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.