Collaborative Research: Fostering Mathematical Modeling Competencies through Collaborative Learning in a Large Language Model (LLM) Simulated Virtual Classroom

Mathematical modeling involves the cognitively demanding process of translating real-life situations into mathematical notations and is a fundamental skill for students pursuing science, technology, engineering, and mathematics (STEM). Cultivating mathematical modeling through collaborative learning can be particularly effective, but orchestrating such collaborative learning tasks requires significant efforts from teachers to oversee group discussions. This could be particularly challenging for marginalized communities that lack teacher resources. This project will explore the development of generative Artificial Intelligence (AI) techniques for creating a virtual classroom platform that supports collaborative learning of mathematical modeling for middle-school students. Through use of the platform the project aims to increase the opportunities for students from under-resourced communities to receive effective mathematics education, supporting more equitable learning. The project will also use the platform as a lens to understand the opportunities and risks of Generative AI techniques and provide insights for future researchers and educators. The virtual classroom platform will include multiple Large Language Model (LLM)-simulated agents/students with which human students can practice collaborative mathematical problem-solving. The project has three research goals, implemented by the interdisciplinary project team with expertise in AI, natural language processing, human-computer interaction, and mathematics education. First, it will address the AI/LLM grounding challenge in the platform development through a neuro-symbolic approach and a modular architecture design. This includes exploring methods for simulated students to behave cohesively in context, aiming to replicate the collaborative behavior of real-life middle-school students during mathematical tasks. Second, it will enhance the platform to serve as an equitable learning environment by conducting participatory design with student users, gathering their input, and leveraging the collected data to refine the platform. Finally, the project involves conducting a series of research studies to understand the efficacy of the platform in fostering students’ mathematical modeling competencies and provide insights into effective ways of applying generative AI in the future of teaching and learning. This project is funded by the “Research on Innovative Technologies for Enhanced Learning (RITEL)” program that supports early-stage exploratory research in emerging technologies for teaching and learning. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.