Playing Dumb to Get Smart: Creating and Evaluating an LLM-based Teachable Agent within University Computer Science Classes

要旨

This work presents the iterative design and evaluation of a large-language-model (LLM) based teachable agent, MatlabTutee, that facilitates learning-by-teaching (LBT) experiences within university computer science courses. We detail four different experiments, with a total of 119 students, where we refine our system, compare it to human-facilitated LBT experiences, and deploy it in two, month-long in-the-wild environments. We find that our system is able to successfully convey a learner persona similar to a human pretending to be novice while also providing comparable LBT benefits. These benefits include helping students identify areas for improvement, develop a more accurate assessment of their own abilities, and improve their overall attitudes toward computer science. We also explore how students choose to adopt our system into their study habits while situated in real university courses.

著者
Kantwon Rogers
Georgia Institute of Technology, Atlanta, Georgia, United States
Michael Mao. Davis
Georgia Institute of Technology, Atlanta, Georgia, United States
Mallesh Maharana
Georgia Institute of Technology, Atlanta, Georgia, United States
Peter Stanley. Etheredge
Georgia Institute of Technology, Atlanta, Georgia, United States
Sonia Chernova
Georgia Institute of Technology, Atlanta, Georgia, United States
DOI

10.1145/3706598.3713644

論文URL

https://dl.acm.org/doi/10.1145/3706598.3713644

動画

会議: CHI 2025

The ACM CHI Conference on Human Factors in Computing Systems (https://chi2025.acm.org/)

セッション: CS Education and Security

G303
6 件の発表
2025-04-28 20:10:00
2025-04-28 21:40:00
日本語まとめ
読み込み中…