Beyond the Leaderboard: Rethinking Medical Benchmarks for Large Language Models

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Chen, Wenting, Yu, Guo, Cheung, Yiu-Fai, Ding, Meidan, Liu, Jie, Ma, Zizhan, Wang, Wenxuan, Shen, Linlin
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866915964877012992
author Chen, Wenting
Yu, Guo
Cheung, Yiu-Fai
Ding, Meidan
Liu, Jie
Ma, Zizhan
Wang, Wenxuan
Shen, Linlin
author_facet Chen, Wenting
Yu, Guo
Cheung, Yiu-Fai
Ding, Meidan
Liu, Jie
Ma, Zizhan
Wang, Wenxuan
Shen, Linlin
contents Large language models (LLMs) show significant potential in healthcare, prompting numerous benchmarks to evaluate their capabilities. However, concerns persist regarding the reliability of these benchmarks, which often lack clinical fidelity, robust data management, and safety-oriented evaluation metrics. To address these shortcomings, we introduce MedCheck, the first lifecycle-oriented assessment framework specifically designed for medical benchmarks. Our framework deconstructs a benchmark's development into five continuous stages, from design to governance, and provides a comprehensive checklist of 46 medically-tailored criteria. Using MedCheck, we conducted an in-depth empirical evaluation of 53 medical LLM benchmarks. Our analysis uncovers widespread, systemic issues, including a profound disconnect from clinical practice, a crisis of data integrity due to unmitigated contamination risks, and a systematic neglect of safety-critical evaluation dimensions like model robustness and uncertainty awareness. Based on these findings, MedCheck serves as both a diagnostic tool for existing benchmarks and an actionable guideline to foster a more standardized, reliable, and transparent approach to evaluating AI in healthcare.
format Preprint
id arxiv_https___arxiv_org_abs_2508_04325
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Beyond the Leaderboard: Rethinking Medical Benchmarks for Large Language Models
Chen, Wenting
Yu, Guo
Cheung, Yiu-Fai
Ding, Meidan
Liu, Jie
Ma, Zizhan
Wang, Wenxuan
Shen, Linlin
Computation and Language
Artificial Intelligence
Computer Vision and Pattern Recognition
Machine Learning
Multimedia
Large language models (LLMs) show significant potential in healthcare, prompting numerous benchmarks to evaluate their capabilities. However, concerns persist regarding the reliability of these benchmarks, which often lack clinical fidelity, robust data management, and safety-oriented evaluation metrics. To address these shortcomings, we introduce MedCheck, the first lifecycle-oriented assessment framework specifically designed for medical benchmarks. Our framework deconstructs a benchmark's development into five continuous stages, from design to governance, and provides a comprehensive checklist of 46 medically-tailored criteria. Using MedCheck, we conducted an in-depth empirical evaluation of 53 medical LLM benchmarks. Our analysis uncovers widespread, systemic issues, including a profound disconnect from clinical practice, a crisis of data integrity due to unmitigated contamination risks, and a systematic neglect of safety-critical evaluation dimensions like model robustness and uncertainty awareness. Based on these findings, MedCheck serves as both a diagnostic tool for existing benchmarks and an actionable guideline to foster a more standardized, reliable, and transparent approach to evaluating AI in healthcare.
title Beyond the Leaderboard: Rethinking Medical Benchmarks for Large Language Models
topic Computation and Language
Artificial Intelligence
Computer Vision and Pattern Recognition
Machine Learning
Multimedia
url https://arxiv.org/abs/2508.04325