MoveBench：面向大语言模型运动与健康指导能力的评测基准
MoveBench: A Rubric-Based Benchmark for Evaluating Large Language Models in Sports and Health Coaching
AI 导读
MoveBench 是首个专门评测大语言模型在运动与健康指导领域能力的基准，覆盖选择题与开放式问答两类任务，配有强调安全合规、科学性与训练逻辑的评分规则。研究对 15 个大语言模型在 809 道题、5,012 条评分标准上进行了大规模评测，发现各模型在安全意识、结构化训练计划设计和个性化能力上表现差异明显，凸显模型“知道”与“做到”之间的差距。相关资源已开源。
正文 · 原文
Abstract:Despite rapid advances in large language models (LLMs), their deployment in the sports and health domain remains high-risk due to the absence of domain-specific evaluation benchmarks. Existing general-purpose benchmarks (e.g., MMLU and HELM) primarily assess broad knowledge and reasoning capabilities, but fail to capture critical domain requirements such as exercise safety, physiological appropriateness, personalization, and actionable coaching quality. This gap is particularly concerning in fitness and health applications, where incorrect or unsafe recommendations can lead to physical harm. To address this limitation, we introduce MoveBench, the first comprehensive benchmarking framework specifically designed for evaluating LLMs in sports and health coaching. The proposed benchmark integrates heterogeneous task formats, including multiple-choice questions (MCQs) and open-ended question answering (OpenQA), grounded in real-world user scenarios and paired with structured rubrics emphasizing safety compliance, scientific validity, and training logic. Through large-scale evaluation of 15 diverse LLMs on 809 questions with 5,012 rubric-level criteria, our results reveal substantial performance variability, particularly in safety awareness, structured program design, and personalization capability, highlighting a critical gap between knowing and doing. Our work establishes a foundational evaluation standard for sports and health LLM applications and provides a reproducible framework for advancing safe, reliable, and context-aware AI-driven coaching systems. Open resources are available at this https URL.
| Comments: | 2 figures. 140 open questions resources: this https URL |
| Subjects: | Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2610.26185 [cs.AI] |
| (or arXiv:2610.26185v2 [cs.AI] for this version) | |
| https://doi.org/10.48550/arXiv.2610.26185 arXiv-issued DOI via DataCite |
Submission history
From: Yitong Li [view email]
[v1]
Mon, 19 Oct 2026 11:16:18 UTC (195 KB)
[v2]
Mon, 26 Oct 2026 03:44:10 UTC (195 KB)
来源：arXiv cs.AI · arxiv.org