Ethel: A Virtual Teaching Assistant

Abstract

Generative AI has shown potential in solving physics problems and providing feedback on assessments. However, results are sometimes still inaccurate, at the wrong level, or using notations and definitions not appropriate for a particular course. A possible solution is augmenting the prompts with course-specific reference materials. Also, for feedback on homework solutions and grading exams, the problem text and the sample solution or grading rubric can be injected into the prompts. Project Ethel at ETH Zurich aims to construct a virtual teaching assistant using these practices.

References in corpus (6)

• Retrieval-Augmented Generation for Knowledge-Intensive NLP Tasks
• Could an Artificial-Intelligence agent pass an introductory physics course?
• Physics task development of prospective physics teachers using ChatGPT
• How do physics students evaluate artificial intelligence responses on comprehension questions? A study on the perceived scientific accuracy and linguistic quality
• The Impact of AI in Physics Education: A Comprehensive Review from GCSE to University Levels
• Can an AI-tool grade assignments in an introductory physics course?

Cited by in corpus (5)

• Assessing Confidence in AI-Assisted Grading of Physics Exams through Psychometrics: An Exploratory Study
• Performance of ChatGPT on tasks involving physics visual representations: the case of the Brief Electricity and Magnetism Assessment
• Creating a customisable Socratic AI physics tutor
• Exploring Large Language Models (LLMs) through interactive Python activities
• When AI Evaluates Its Own Work: Validating Learner-Initiated, AI-Generated Physics Practice Problems