节点文献

生成式人工智能在我国研究生学术写作中的应用现状及其影响

The Application Status and Impact of Artificial Intelligence Generated Content in Postgraduate Academic Writing in China

  • 推荐 CAJ下载
  • PDF下载
  • 不支持迅雷等下载工具,请取消加速工具后下载。

【作者】 蔡芬贾枭沈文钦

【Author】 CAI Fen;JIA Xiao;SHEN Wenqin;Peking University;

【机构】 北京大学全国医学教育发展中心/教育学院北京大学教育学院

【摘要】 随着生成式人工智能(AIGC)工具在学术写作领域的广泛应用,我国研究生群体所受的影响亟待关注。对14 817份研究生问卷数据进行分析发现:相比于硕士研究生,博士研究生对使用AIGC工具辅助学术写作的需求更强;“双一流”建设高校、工学、高年级以及求学外在动机强、同伴功利发表氛围感知强的研究生,更有可能使用AIGC工具辅助写作。研究生主要利用AIGC工具辅助搜索与研究主题相关的资料、翻译/润色论文语言表述,将AIGC工具用于辅助撰写课程作业、翻译/润色论文对研究生的科研学习效率有积极影响,但用于辅助研究设计可能会阻碍其学术写作能力的提升。据此,建议高校优化学术评价环境,加强针对AIGC工具的学术伦理规范建设和使用技能训练。

【Abstract】 With the widespread use of Artificial Intelligence Generated Content(AIGC) tools in academic writing, the impact on postgraduate students in China warrants attention. An analysis of survey data from 14 817postgraduates reveals that doctoral students have a stronger demand for using AIGC tools to assist in academic writing compared to master’s students. Postgraduates from “Double First-Class” construction universities, those studying in engineering, higher academic years, and those with extrinsic motivations for pursuing studies or perceiving a strong utilitarian publication atmosphere among peers are more likely to use AIGC tools for writing assistance. Postgraduates primarily utilize AIGC tools to assist in searching for materials related to their research topics and translating/refining the language expression in their papers. Postgraduates who use AIGC tools for coursework, translation, or refinement are more likely to recognize the positive impact of these tools on their research efficiency, but their use in research design may hinder the development of academic writing skills. Based on these findings, it is recommended that universities improve academic evaluation standards, and strengthen the development of academic ethical norms and usage skill training for AIGC tools.

【基金】 2023年度国家自然科学基金青年科学基金项目“基于IPOD模型的‘医+X’博士生跨学科培养模式及其成效研究”(72304019)的研究成果
  • 【文献出处】 中国高教研究 ,China Higher Education Research , 编辑部邮箱 ,2025年01期
  • 【分类号】TP18;G643
  • 【下载频次】852
节点文献中: 

本文链接的文献网络图示:

本文的引文网络