Enhancing Undergraduate STEM Education with Large Language Models: Personalization, Collaboration, and Active Learning

About This Grant

This Level 1 IUSE Engaged Student Learning project from the University of Massachusetts Lowell serves the national interest by developing AI-driven educational technology tools to support student learning in undergraduate computer science courses. Specifically, this project will explore how large-language models (LLMs) can be customized to enable tailored, interactive, and reflective learning experiences. The project-developed LLM will draw from student notes and course materials in tandem with existing data sets to personalize their learning experiences. The project tool will be iteratively improved through a collaborative process driven by students and will be used to simulate virtual students that will prompt users and encourage active learning. This project will explore how LLM tools can be better developed to support individualized learning and reflection. The project goals are: (1) to examine how LLMs can provide personalized, content-specific support for STEM students; (2) to examine how LLMs can facilitate collaborative learning environments; and (3) to investigate how LLMs can support the design of active learning experiences. The project will use retrieval-augmented generation techniques to align LLM outputs with course materials and student work to build engagement and reduce the incidence of generic responses. The project will generate knowledge through a rigorous evaluation plan that will utilize a mixed-methods approach to explore the LLM development and refinement process and capture impacts on student users. The NSF IUSE: EDU Program supports research and development projects to improve the effectiveness of STEM education for all students. Through the Engaged Student Learning track, the program supports the creation, exploration, and implementation of promising practices and tools. 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.