26–30 Aug 2024
Auditorium Maximum UJ
Europe/Warsaw timezone

Practical evaluation of the possibilities of integrating Large Language Models in physics laboratory instruction

30 Aug 2024, 11:30
20m
Small Hall

Small Hall

Oral presentations Multimedia, AR, VR and AI in Physics Education Oral presentations

Speaker

Marina Babayeva (Charles University)

Description

A team of researchers from Portland State University (PSU) and Prague’s Charles University pioneered the integration of OpenAI's GPT-4 Large Language Module (LLM) into the introductory physics laboratory, serving as a virtual teaching assistant. A class of 26 students participated in a pilot implementation of the AI assistant for introductory physics at PSU. We evaluated students’ interactions with the chatbot, assessed the quality of the LLM’s responses, and conducted surveys to gather qualitative insights from students’ experiences and attitudes. This contribution presents the key findings of our pilot study, including examples from the transcripts and surveys.

How would you like to present your contribution? Live in Kraków (time slot to be allotted based on the programme)
Target education level University
Category Formal Education

Authors

Marina Babayeva (Charles University) Dr Ralf Widenhorn (Portland State University) Mr Travis Kregear (Portland State University)

Presentation materials