生成式人工智能赋能电子信息研究生课程教学改革

Teaching Reform of Graduate Courses in Electronic Information Empowered by Generative Artificial Intelligence

摘要: 生成式人工智能(Generative Artificial Intelligence, GenAI)的迅猛发展为高等教育课程改革提供了新的技术条件与方法范式。面向电子信息专业研究生培养中占据核心地位的高级软件工程类课程长期存在内容更新滞后、真实情境案例匮乏、教学重知识轻能力、与行业需求脱节等共性问题。本文聚焦GenAI在课程目标重构、教学资源生产、个性化学习支持与全过程评价中的系统嵌入，提出以“智能生成课程资源”和“面向能力的个性化培养”双核驱动的教学设计框架。在该框架下，利用人工智能生成内容(Artificial Intelligence Generated Content, AIGC)工具实现项目化案例与需求文档的半自动生产、代码示例与实验框架的协同生成与迭代更新，同时以能力要素矩阵为牵引构建分层进阶的学习任务链与形成性评价机制，从而打通“问题–分析–实现–验证–迭代”的工程闭环。研究旨在为电子信息类研究生课程的智能化转型与高质量发展提供可迁移、可落地的设计思路与操作路径。

Abstract: The rapid development of Generative Artificial Intelligence (GenAI) has provided new technological conditions and methodological paradigms for curriculum reform in higher education. In the cultivation of graduate students majoring in electronic information, advanced software engineering courses occupy a central position; however, they have long faced common challenges such as delayed content updates, a lack of authentic real-world scenarios, an emphasis on knowledge transmission over competence development, and misalignment with industry demands. This paper focuses on the systematic integration of GenAI into course objective reconstruction, instructional resource generation, personalized learning support, and whole-process assessment. A teaching design framework driven by two core components—intelligent generation of course resources and competence-oriented personalized cultivation—is proposed. Within this framework, Artificial Intelligence Generated Content (AIGC) tools are employed to enable semi-automatic generation of project-based cases and requirements documents, collaborative generation and iterative updating of code examples and experimental frameworks, and continuous enrichment of teaching resources. Meanwhile, guided by a competence-element matrix, a hierarchical and progressive chain of learning tasks and a formative assessment mechanism are constructed, thereby establishing an integrated engineering closed loop of “problem identification-analysis-implementation-verification-iteration”. This study aims to provide transferable and practically applicable design concepts and implementation pathways for the intelligent transformation and high-quality development of graduate-level courses in electronic information disciplines.