A comprehensive qualitative evaluation framework for large language models (LLM) in healthcare that expands beyond traditional accuracy and quantitative metrics needed. We propose 5 key aspects for evaluation of LLMs: Safety, Consensus, Objectivity, Reproducibility and Explainability (S.C.O.R.E.). We suggest that S.C.O.R.E. may form the basis for an evaluation framework for future LLM-based models that are safe, reliable, trustworthy, and ethical for healthcare and clinical applications.
@article{arxiv.2407.07666,
title = {A Proposed S.C.O.R.E. Evaluation Framework for Large Language Models : Safety, Consensus, Objectivity, Reproducibility and Explainability},
author = {Ting Fang Tan and Kabilan Elangovan and Jasmine Ong and Nigam Shah and Joseph Sung and Tien Yin Wong and Lan Xue and Nan Liu and Haibo Wang and Chang Fu Kuo and Simon Chesterman and Zee Kin Yeong and Daniel SW Ting},
journal= {arXiv preprint arXiv:2407.07666},
year = {2024}
}