AI-Powered Mathematics Education Project Funded by UF President’s Strategic Funding Initiative

Dr. Wanli Xing

A new collaborative research project will use AI to facilitate mathematics education in an interactive and unique way. “SALT-Math: Scalable AI-Augmented Learning by Teaching for Math Education” aims to revolutionize K-12 mathematics learning by implementing a learning-by-teaching framework that uses a large language model to flip students’ roles as teachers to AI agents (resulting in enhanced student outcomes). This project has received $930,000 in funding over three years.

SALT-Math is one of 10 new projects that received funding through University of Florida President Ben Sasse’s strategic funding initiative. The funding initiative is designed to advance interdisciplinary scholarship and enhance the student experience.

Dr. My T. Thai

This project is a collaborative effort between the Warren B. Nelms Institute for the Connected World and the Lastinger Center for Learning. Researchers working on this project are Dr. Wanli Xing (Director of Education Programs, Warren B. Nelms Institute), Dr. Phil Poekert (Director, Lastinger Center for Learning), Dr. Zandra de Araujo, (Mathematics Principal, Lastinger Center for Learning) and Dr. My T. Thai (Associate Director, Warren B. Nelms Institute).

SALT-Math will be created and distributed in the context of the Lastinger Center for Learning’s Math Nation platform, which was adopted as the core K-12 math curriculum in the state of Florida (and engages more than 1 million K-12 students across the United States annually).  


About the Project

SALT-Math aims to revolutionize the math learning process by implementing a learning-by-teaching (LT) framework using socially responsible LLMs to flip students’ role as teachers to AI agents, resulting in substantially enhanced students’ motivation, engagement, and algebra learning outcomes. In essence, SALT-Math is an LLM-powered AI system that simulates fictional students with realistic educational needs, then asks real students to tutor them. The highlighted originality of SALT-Math is that it catalyzes a paradigm shift, moving beyond the current focus on constructing superior AI to guide learning, and embracing the potential of interactive and enjoyable AI that motivates and engages students. This project strives to make algebra learning a fun and proactive experience for students to deepen their understanding of algebra concepts. After all, while we are training LLMs to be BETTER TEACHERS (or TUTORS), they are already EXCEPTIONAL STUDENTS. Given the project’s scope, our primary focus will be on developing SALT-Math for Algebra I, an area that is likely of high demand.