基于大语言模型的课程教学答疑系统设计

Design of Course Teaching Q&A System Based on Large Language Model

摘要: 在混合式教学中，如何能对学生学习过程中的疑问进行及时回答对教师来说是一个比较头痛的问题。借助AI技术，开发能自动回复学生问题的教学辅助系统是一个非常重要手段。本文介绍一个能协助教师，帮助学生解决课程学习问题的会话系统ChatQA。该系统通过深度学习模型，能够与学生进行自然、连贯的对话，回答学生课程学习中的问题，有效缓解了教师的答疑负担。论文详细讨论了该系统的设计原理、Prompt优化方法等，并通过软件工程课程的问题进行了测试，显示了该方法的可用性，具有较好的自动答疑的效果。

Abstract: In blended teaching, how to answer students’ questions in the learning process in a timely manner is a headache for teachers. With the help of AI technology, it is very important to develop a teaching assistance system that can automatically respond to students’ questions. This article introduces ChatQA, a conversational system that can help teachers and students solve course learning problems. Through the deep learning model, the system can have a natural and coherent dialogue with students, answer students’ questions in course learning, and effectively alleviate the burden of answering questions on teachers. The design principle and prompt optimization method of the system are discussed in detail, and the questions of the software engineering course are tested, which shows the usability of the method and has a good effect on automatic Q&A.