Large language models (LLMs) show promise for supporting clinical decision-making in complex fields such as rheumatology. Our evaluation shows that smaller language models (SLMs), combined with retrieval-augmented generation (RAG), achieve higher diagnostic and therapeutic performance than larger models, while requiring substantially less energy and enabling cost-efficient, local deployment. These features are attractive for resource-limited healthcare. However, expert oversight remains essential, as no model consistently reached specialist-level accuracy in rheumatology.
@article{arxiv.2507.07983,
title = {Performance and Practical Considerations of Large and Small Language Models in Clinical Decision Support in Rheumatology},
author = {Sabine Felde and Rüdiger Buchkremer and Gamal Chehab and Christian Thielscher and Jörg HW Distler and Matthias Schneider and Jutta G. Richter},
journal= {arXiv preprint arXiv:2507.07983},
year = {2025}
}