Simulating Classroom Education with LLM-Empowered Agents

• DOI: 10.48550/arxiv.2406.19226
• Publication Date: 2024-06-27
• Authors: Zheyuan Zhang; Daniel Zhang-Li; Jifan Yu; Linlu Gong; Jinchang Zhou; Zhiyuan Liu; Lei Hou; Juanzi Li
• Keywords: FOS: Computer and information sciences; Computer Science - Computation and Language; Computer Science - Human-Computer Interaction; Computation and Language (cs.CL); Human-Computer Interaction (cs.HC)

Large language models (LLMs) have been employed in various intelligent educational tasks to assist teaching. While preliminary explorations focused on independent LLM-empowered agents for specific tasks, the potential LLMs within a multi-agent collaborative framework simulate classroom with real user participation remains unexplored. In this work, we propose SimClass, simulation involving participation. We recognize representative class roles and introduce novel control mechanism automatic teaching, conduct experiments two real-world courses. Utilizing Flanders Interactive Analysis System Community of Inquiry theoretical frame works from analysis, demonstrate that can traditional interaction patterns effectively while enhancing user's experience. also observe emergent group behaviors among where collaborate create enlivening interactions classrooms improve learning process. hope work pioneers application systems virtual