RAGCare-QA: A Benchmark Dataset for Evaluating Retrieval-Augmented Generation Pipelines in Theoretical Medical Knowledge
Listed in
This article is not in any list yet, why not save it to one of your lists.Abstract
The paper introduces RAGCare-QA, an extensive dataset of 420 theoretical medical knowledge questions for assessing Retrieval-Augmented Generation (RAG) pipelines in medical education and evaluation settings. The dataset includes one-choice-only questions from six medical specialties (Cardiology, Endocrinology, Gastroenterology, Family Medicine, Oncology, and Neurology) with three levels of complexity (Basic, Intermediate, and Advanced). Each question is accompanied by the best fit of RAG implementation complexity level, such as Basic RAG (315 questions, 75.0%), Multi-vector RAG (82 questions, 19.5%), and Graph-enhanced RAG (23 questions, 5.5%). The questions emphasize theoretical medical knowledge on fundamental concepts, pathophysiology, diagnostic criteria, and treatment principles important in medical education. The dataset is a useful tool for the assessment of RAG-based medical education systems, allowing researchers to fine-tune retrieval methods for various categories of theoretical medical knowledge questions.
VALUE OF THE DATA
RAGCare-QA dataset is designed to benchmark state-of-the-art RAG architectures recommendations for theoretical medical knowledge through 420 human annotated single-choice questions, well-distributed in 6 different medical specialties.
Researchers can leverage this resource to build more effective educational tools that adapt their retrieval strategies based on question complexity and medical specialty.
The dataset fills a gap in medical AI by providing a standardized benchmark that supports the development of AI-based adaptive educational tools.
The dataset classifies each question by the most suitable RAG architecture, Basic, Multi-vector, or Graph-enhanced, needed for context retrieval, enabling precise performance comparisons across retrieval strategies.
The dataset can serve as a foundation for development of specialized retrieval strategies to enhance learning outcomes in medical education.
