From substitution to human-AI collaboration: a case study on assessment of reflective writing assignments of medical students using reasoning-capable large language models
Digital Intelligence MedicalEducation|更新时间:2026-02-13
|
From substitution to human-AI collaboration: a case study on assessment of reflective writing assignments of medical students using reasoning-capable large language models
Chinese Journal of Medical Education ResearchVol. 25, Issue 1, Pages: 1-11(2026)
作者机构:
1.四川大学华西第二医院筑浪学院,成都 610041
2.国家教材建设重点研究基地(高等学校医学教材研究),成都 610041
3.四川大学华西医学中心医学教学部,成都 610041
作者简介:
Yan Yujiang, Email: 1051173813@qq.com
基金信息:
2025 Research Project of Medical Textbook Development of Sichuan University (Major Project)(SCUYJ0126);The 11th Higher Educational Teaching Reform Project of Sichuan University(SCU11404;SCU11411);2025 Research Project of West China Center for Research and Development of Women and Children's Health(XBFY-YJXM2025003)
Chai Hua, Yan Yujiang, Qing Ping. From substitution to human-AI collaboration: a case study on assessment of reflective writing assignments of medical students using reasoning-capable large language models[J]. Chinese Journal of Medical Education Research, 2026, 25(1): 1-11.
Chai Hua, Yan Yujiang, Qing Ping. From substitution to human-AI collaboration: a case study on assessment of reflective writing assignments of medical students using reasoning-capable large language models[J]. Chinese Journal of Medical Education Research, 2026, 25(1): 1-11. DOI: 10.3760/cma.j.cn116021-20250805-02199.
From substitution to human-AI collaboration: a case study on assessment of reflective writing assignments of medical students using reasoning-capable large language models
本研究对具有深度推理功能的大语言模型(reasoning-capable large language models, rcLLMs)评阅医学生反思性写作作业的可行性和效果进行了探索。在选择作业样本、建立评阅方法的基础上,用2种国内主流rcLLMs进行评阅,并采用定量与定性相结合的混合研究方法将评阅结果与教师评阅结果及评分标准相比较。研究发现2种rcLLMs和教师三者评分的一致性及两两之间评分的相关性较差,2种rcLLMs各自评分的稳定性也较差;rcLLMs生成的评语结构完整,但关注主题与评分标准不完全一致。本研究揭示出目前rcLLMs尚不能作为评阅医学生反思性写作作业的有效工具,讨论了相关影响因素,并强调了建立人机协作策略及反思AI评阅正当性的重要性。
Abstract
This case study explored the feasibility and efficacy of employing reasoning-capable large language models (rcLLMs) to assess the reflective writing assignments of medical students. Following the selection of writing samples and the establishment of an assessment methodology
two prominent domestic rcLLMs were utilized to evaluate the assignments. A method combining quantitative and qualitative analyses was employed to compare the evaluations from rcLLMs with those from instructors in accordance with the established grading rubric. The findings revealed poor inter-rater reliability due to weak correlations among all three raters collectively and pairwise. Both rcLLMs demonstrated significant variability in their own scoring consistency. While the feedback comments generated by the rcLLMs were well-structured
their focuses exhibited a misalignment with the grading rubric. This study suggests that current rcLLMs are not yet suitable for use as effective tools for assessing the reflective writing assignments of medical students. The influencing factors were discussed
and the need to develop human-AI collaborative strategies
as well as to critically reflect on the legitimacy of employing rcLLMs in assessment
DeepSeek-AI , Guo D , Yang D , et al . DeepSeek-R1: incentivizing reasoning capability in LLMs via reinforcement learning [J ] . arXiv , 2025 : abs/2501.12948. DOI: 10.48550/arXiv.2501.12948 http://dx.doi.org/10.48550/arXiv.2501.12948 .
OPEN AI . ChatGPT: Optimizing Language Models for Dialogue [EB/OL ] . ( 2022-11-30 ) [ 2025-07-31 ] . https://openai.com/index/chatgpt/ https://openai.com/index/chatgpt/ .
Kimi Team , Du A , Gao B , et al . Kimi k1.5: scaling reinforcement learning with LLMs [J ] . arXiv , 2025 : abs/2501.12599. DOI: 10.48550/arXiv.2501.12599 http://dx.doi.org/10.48550/arXiv.2501.12599 .
Li J , Miao B , Wang S , et al . Hiplot: a comprehensive and easy-to-use web service for boosting publication-ready biomedical data visualization [J ] . Brief Bioinform , 2022 , 23 ( 4 ): bbac261 . DOI: 10.1093/bib/bbac261 http://dx.doi.org/10.1093/bib/bbac261 .
Gao C , Wu X , Lin Z , et al . NExtLong: toward effective long-context training without long documents [J ] . arXiv , 2025 : abs/2501.12766. DOI: 10.48550/arXiv.2501.12766 http://dx.doi.org/10.48550/arXiv.2501.12766 .
Xiong W , Liu J , Molybog I , et al . Effective long-context scaling of foundation models [J ] . arXiv , 2023 : abs/2309.16039v 3 . DOI: 10.48550/arXiv.2309.16039 http://dx.doi.org/10.48550/arXiv.2309.16039 .
Gao C , Wu X , Fu Q , et al . Quest: query-centric data synthesis approach for long-context scaling of large language model [J ] . arXiv , 2024 : abs/2405.19846. DOI: 10.48550/arXiv.2405.19846 http://dx.doi.org/10.48550/arXiv.2405.19846 .
Practice and exploration of adolescent medicine education with an international perspective: a case study of Zhejiang University Sustainable Development Global Summer School
A qualitative study on the experience of publishing case reports by resident physicians: challenges, strategies, and expectations
Promoting deep integration of artificial intelligence and education for new pathways of precision learning
Research progress and prospects of artificial intelligence in new models of clinical reasoning education
Design of Novel High Accuracy Fine Tuning Device for Optical Fiber
Related Author
Lao Yanyun
Zhang Ling
Sun Xiaoying
Zhang Li
Cheng Xiaoying
Fu Junfen
Yang Zhihui
Chen Wen
Related Institution
International Affairs Office, Children's Hospital, School of Pediatrics, Zhejiang University School of Medicine
Department of Dermatology and Venerology, Peking University First Hospital/National Clinical Research Center for Skin and Immune Diseases/Beijing Key Laboratory of Molecular Diagnosis of Dermatoses, Peking University First Hospital/NMPA Key Laboratory for Quality Control and Evaluation of Cosmetics, Peking University First Hospital
Department of General Practice, Peking University First Hospital
China Medical University
Department of General Internal Medicine, Peking Union Medical College Hospital, Chinese Academy of Medical Science & Peking Union Medical College