Three Seed Grants Awarded for AI and Education Research

Carnegie Mellon University (CMU) is at the forefront of integrating technology into education, offering tools and expertise that benefit both its own community and the broader field of education. In a bid to advance generative AI in educational settings, CMU initiated a seed grant program last year aimed at supporting faculty in developing innovative AI tools. This year, the program has selected three faculty-led initiatives dedicated to enhancing teaching and learning through generative AI.

Each project can receive up to $150,000 in funding, a collaborative effort between CMU and its donors, aimed at pioneering research that could significantly alter educational practices and outcomes.

Supporting Mastery in Student Learning

One of the highlighted projects is led by Paulo Carvalho, an assistant professor in CMU’s Human-Computer Interaction Institute. Carvalho is exploring how generative AI can enhance student engagement and enable more personalized education. His research focuses on creating environments that prioritize active practice using data analytics and computational modeling to refine student learning experiences.

Paulo Carvalho, joined by HCII postdoctoral fellow Michael Asher, is developing a feedback tool centered around mastery-based learning. This tool aims to provide students with personalized feedback on open-ended questions, promoting deeper cognitive engagement beyond traditional multiple-choice formats.

Designing Effective Feedback Frameworks

John Stamper, an associate professor at the Human-Computer Interaction Institute and technical director of LearnSphere’s DataShop, is another grant recipient. Stamper’s project focuses on developing a framework for crafting feedback prompts powered by large language models (LLMs), guided by principles from the learning sciences.

This initiative leverages the Langchain library, offering API access to numerous LLMs, and aims to integrate this framework into platforms such as CMU’s Open Learning Initiative. The goal is to equip educators with tools to generate LLM-driven feedback, aiding instructional and research efforts by testing various feedback types in different educational settings.

Enhancing Learning Through Immersive Experiences

The third project, led by Param Vir Singh, associate dean for research at CMU’s Tepper School of Business, seeks to create an interactive case analysis tool using AI. This tool will allow students to engage with dynamic scenarios that evolve based on their decisions, rather than following a predetermined path.

This project team includes Laurence Ales, professor of economics, John Gasper, associate teaching professor of economics, and AI project engineer Bambi Brewer. Their objective is to provide students with an adaptive learning experience that mirrors real-world complexities, initially within the Tepper School with potential expansion to other fields.