题名：基于大模型的教学智能体构建与应用研究
Title: Research on the Construction and Application of Teaching Agents Based on Large Models
作者：卢 宇，余京蕾，陈鹏鹤
Authors: Lu Yu, Yu Jinglei, Chen Penghè
摘要：随着生成式人工智能的快速发展，基于大模型的智能体已经逐步具备了多模态感知、检索增强生成、推理与规划、交互与进化等能力。该研究提出基于大模型的教学智能体的基本概念与框架，以“大模型”为技术核心，重点构建“教育任务设定”“教育任务规划”“教育能力实现与拓展”“教育内容记忆与反思”“交互协作与动态进化”多个功能模块，支持与多类型对象交互并实现动态进化，涵盖人机交互、多智能体交互以及环境交互。基于所提出的框架，研究以项目式学习任务为应用场景，阐述了教学智能体作为“助教智能体”和“同伴智能体”，在个性化驱动问题提出、项目方案共同设计、项目作品协作完成、项目作品多角色评价多个环节的作用及相关支撑技术。最后，研究进一步探讨了教学智能体的发展方向与未来展望。
Abstract: With the rapid development of generative artificial intelligence, agents based on large models have gradually acquired capabilities such as multimodal perception, retrieval-augmented generation, reasoning and planning, interaction and evolution. This research puts forward the basic concept and framework of teaching agents based on large models. Taking "large models" as the core technology, it focuses on constructing multiple functional modules including "educational task setting", "educational task planning", "realization and expansion of educational capabilities", "memory and reflection of educational content", and "interactive collaboration and dynamic evolution", which support interaction with multiple types of objects and achieve dynamic evolution, covering human-computer interaction, multi-agent interaction and environmental interaction. Based on the proposed framework, taking project-based learning tasks as the application scenario, this research elaborates on the roles and related supporting technologies of teaching agents as "teaching assistant agents" and "peer agents" in multiple aspects such as the proposal of personalized-driven questions, the joint design of project plans, the collaborative completion of project works, and the multi-role evaluation of project works. Finally, the research further explores the development direction and future prospects of teaching agents.
关键词：教学智能体；大模型；生成式人工智能；项目式学习
Keywords: Teaching agents; Large models; Generative artificial intelligence; Project-based learning