Collaborative Research: Phase I CAMEL-CN: ALTA-CN: AI-Powered Learning by Teaching for Algebraic Reasoning through Collaborative Networks

This project leverages existing data, which include millions of interactions between middle school students and Artificial Intelligence (AI) pedagogical agents, annotated tutoring logs, and linked statewide mathematics assessments, to create research-ready datasets. The project is well-positioned to expand national research infrastructure by developing and releasing large-scale, well-documented, and AI-ready datasets mapped to validated mathematics assessment outcomes. The outcomes of this project can be used to inform research on mathematics teaching and learning. The project leverages an established digital learning platform used by over 60,000 middle school students across the country. This platform takes a novel approach to promote mathematics learning by positioning students as active instructors who articulate reasoning, diagnose errors, and refine explanations as they teach mathematics concepts to an AI agent. This ‘learning-by-teaching’ approach is well grounded in the science of learning and cognitive science research but is novel in the space of personalized digital learning. The dataset includes over 3 million student-AI learning-by-teaching interactions and over 30 million learning engagement events, such as instructional video watching and question-answer interactions. The research team aims to connect these rich records of student engagement and learning to high stakes assessments of mathematics outcomes in middle school. The project advances our understanding of AI-powered mathematics learning through the creation, documentation, and dissemination of large-scale, multimodal learning-by teaching datasets. 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. NSF Award ID: 2621009 | Program: 01002526DB NSF RESEARCH & RELATED ACTIVIT | Principal Investigator: Chenglu Li | Institution: University of Utah, SALT LAKE CITY, UT | Award Amount: $999,998 View on NSF Award Search: https://www.nsf.gov/awardsearch/show-award/?AWD_ID=2621009 View on Research.gov: https://www.research.gov/awardapi-service/v1/awards/2621009.html