用AI模拟医患角色，比真人更稳定且省钱。
Human or LLM as Standardized Patients? A Comparative Study for Medical Education
- 分模块设计让AI患者回答更自然可控
- 实验显示其表现接近真人，新手学得更快
- 适合医学教育机构降本增效
标准化患者（SP）在临床技能训练中不可或缺，但成本高且难扩展。尽管已有基于大语言模型（LLM）的虚拟标准化患者（VSP），但其行为不稳定，缺乏与真人对比。我们提出EasyMED多智能体VSP框架，将病例信息披露与响应生成分离，实现稳定、依询问变化的患者行为。同时构建SPBench基准，包含八项专家定义的交互评估标准。实验表明，EasyMED在案例一致性与受控披露方面优于现有VSP，更接近真人表现。四周期对照研究显示，学习效果与真人训练相当，尤其对初学者提升更明显，且具更强灵活性、心理安全感和成本效益。
原文摘要 · Abstract (English)
Standardized patients (SPs) are indispensable for clinical skills training but remain expensive and difficult to scale. Although large language model (LLM)-based virtual standardized patients (VSPs) have been proposed as an alternative, their behavior remains unstable and lacks rigorous comparison with human standardized patients. We propose EasyMED, a multi-agent VSP framework that separates case-grounded information disclosure from response generation to support stable, inquiry-conditioned patient behavior. We also introduce SPBench, a human-grounded benchmark with eight expert-defined criteria for interaction-level evaluation. Experiments show that EasyMED more closely matches human SP behavior than existing VSPs, particularly in case consistency and controlled disclosure. A four-week controlled study further demonstrates learning outcomes comparable to human SP training, with stronger early gains for novice learners and improved flexibility, psychological safety, and cost efficiency.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。