Leveraging Large Language Models to Generate 3D Simulations of Complex Course Concepts in a Metaverse Environment

The proposed work aims to integrate the metaverse into the preparation of computer science materials for education. Students often struggle with engagement, motivation, and conceptual ambiguity, especially when dealing with abstract, theory-based learning materials. To address these challenges, we propose deploying Metaverse technologies and Large Language Models (LLMs) to generate 3D simulation-based course examples that clarify complex algorithm executions. The goal is to assist novice teachers in preparing concrete, real-world examples for ambiguous concepts. To this end, we have developed an LLM-powered tool that supports teachers in two key steps: first, generating abstract examples that relate a given concept to real-world applications. Second, transforming these abstract examples into concrete, interactive 3D simulations, enabling students to better understand abstract concepts, making them more tangible and engaging. By integrating metaverse-based simulations with AI-assisted content generation, our approach aims to enhance students’ comprehension, engagement, and problem-solving skills in computer science education. The obtained results are very promising.