Body
Image
banner that has 2024 ITEST Meeting name on it

Embedding LLM-powered conversational tutors in STEM videos through teacher-AI collaboration

The rapid advances in LLMs present tremendous opportunities to create interactive, personalized learning experiences at scale. However, to truly harness the educational potential of these technologies, it's crucial that teachers - who are at the forefront of daily student interaction and possess indispensable knowledge and expertise - go beyond being mere consumers to key contributors. This one-year project proposes a teacher-AI collaboration paradigm that can potentially realize this vision. In this project, we partner with WGBH Educational Foundation (GBH) to support teachers to create and customize intelligent tutor accompanying the STEM learning resources available on PBS LearningMedia. We have performed an interview study with high school science teachers to understand how they teach videos in their lessons. Teachers overwhelmingly liked engaging students in a conversation about the video as a method for inquiry and learning. The interview study shed light on the tutoring strategies teachers would use. For example, teachers make sure the questions do not go beyond the scope of the video, and they do not probe further when the students begin to get frustrated. Based on the findings, we are developing a platform that enables teachers to configure LLM-powered conversational tutors for STEM videos. We employ LLM alignment techniques and human-AI collaborative techniques to ensure that the conversational tutors provide a safe and desirable learning experience to students.

Pillar 1: Innovative Use of Technologies in Learning and Teaching

This project aims to embed conversational tutors to help students learn videos, which is an innovative learning experience for students. Moreover, this project innovates teacher-AI collaborative mechanisms in supporting teachers to configure and customize dynamic tutoring experience for students’ interaction with AI. This project will experiment with and combine LLM alignment techniques, RAG techniques, multi-agent approaches, and human-AI collaborative techniques in novel ways.

Pillar 2: Partnerships for Career and Workforce Preparation.

In this project, we partner with WGBH Educational Foundation (GBH), Boston’s PBS station, to support teachers to create and customize intelligent tutor accompanying the STEM learning resources available on PBS LearningMedia, a platform already popular among teachers nationwide, for its comprehensive library of over 5,000 high quality STEM-focused videos and millions of users. Our prototype will embed conversational tutors in STEM videos on PBS LearningMedia.

Pillar 3: Strategies for Equity in STEM Education

First, the conversational tutors would in particular benefit students who have limited access to expert teachers. One of the aims of this project is to provide personalized learning experiences at scale with the conversational tutors. Second, the conversational experience provides scaffolding during the students’ question answering process. This may in particular help the students who do not do well in traditional problem-solving or question answering activities.
This image describes a system design for the proposed teacher interface, where teachers see AI-suggested answer components and key concepts and can make edits. The system takes minimal user input and will provide desirable interaction experience to students.
Discipline(s)
Environmental sciences
Target Gradespan(s)
High school (9-12)
Target Participant(s)
Educators
Project Setting(s)
Formal Education
Category
RAPID
Body

 

2024 ONLINE PROGRAM

Visit the 2024 ITEST PI Meeting Online Program