发表机构
McGill University; Mila – Quebec AI Institute; Institute of Health Sciences Education, McGill University; School of Social Work, McGill University; Montreal Children’s Hospital, McGill University Health Centre; Steinberg Centre for Simulation and Interactive Learning, McGill University; Linarite AI（麦吉尔大学; 米拉-魁北克人工智能研究所; 麦吉尔大学健康科学教育研究所; 麦吉尔大学社会工作学院; 麦吉尔大学健康中心蒙特利尔儿童医院; 麦吉尔大学斯坦伯格模拟与互动学习中心; 利纳里特人工智能公司）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
针对儿科重症沟通（SIC）培训的需求，研究推出首个相关基准套件与自改进模拟器框架SIC-Agents，其性能优于静态专家提示，且发布了相关基准供未来研究使用。
AI 中文摘要
儿科重症沟通（SIC）至关重要，但临床医生的可扩展沟通培训仍十分有限。与其他对话模拟场景相比，儿科SIC面临更多挑战，包括多方互动、应对家长情绪困扰以及对反馈动态的强依赖性。现有基于大语言模型（LLM）的模拟器优化的是通用对话质量，而非有效SIC培训所需的课程相关行为。我们与教育工作者和儿科临床医生合作，推出首个针对儿科SIC培训的基准套件和模拟框架。我们的基准PitfallBench和DialogueBench分别从对话轮次级别和完整对话层面评估模拟器。我们进一步提出SIC-Agents，这是一种自改进框架，可生成临床医生可编辑的技能文档以指导模拟器行为。实验表明，SIC-Agents的性能优于静态专家提示。为支持未来研究，我们在该httpsURL发布了用于儿科SIC中家长模拟的基准。
英文摘要
Pediatric serious illness communication (SIC) is critically important, yet scalable communication training for clinicians remains limited. Compared with other dialogue simulation settings, pediatric SIC poses additional challenges, including multi-party interactions, response to parental distress and strong dependence on feedback dynamics. Existing LLM-based simulators optimize generic dialogue quality rather than curriculum-contingent behavior required for effective SIC training. In collaboration with educators and pediatric clinicians, we introduce the first benchmark suite and simulation framework tailored to pediatric SIC training. Our benchmarks, PitfallBench and DialogueBench, evaluate simulators both at the turn-level and across full dialogues. We further propose SIC-Agents, a self-improving framework that generates a clinician-editable skill document to guide simulator behavior. Our experiments show that SIC-Agents outperforms static expert prompting. To support future research, we release our benchmarks for parent simulation in pediatric SIC at https://github.com/Beikewzh/sic-benchmarks
CommentsAccepted to EMNLP 2026
