Integrating LLMs in Higher Education, Through Interactive Problem Solving and Tutoring: Algorithmic Approach and Use Cases
摘要
Despite the concerns that recent developments in Large Language Models (LLMs) have raised, they undoubtedly revealed a novel potential of Artificial Intelligence (AI) algorithms in educational environments. Whether they are used for tutoring, in a manner similar to that of Intelligent Tutoring Systems (ITS), or to support assessment design and delivery, their impact in a learning setting is remarkable. In this paper, we propose an interactive tutoring approach, utilizing ChatGPT’s API. By exploiting ChatGPT’s programming interface, we can develop customized interactive problem-solving and tutoring sessions on specific topics of interest. The API’s versatility allows for dynamic interactions, fostering a deeper understanding of subjects taught and effective problem-solving skills. We demonstrate the application of the developed code in an applied educational setting with specific use cases.