Course Description
Examination of human–artificial intelligence interaction (AI) in educational assessment, including the use of large language models for assessment development and scoring. Topics include prompt engineering, fine-tuning, human-AI collaboration, human-in-the-loop supervision, and issues of validity, fairness, transparency, and trust in AI-generated outputs.
Athena Title
Human-AI Interaction
Semester Course Offered
Offered spring
Grading System
A - F (Traditional)
Student learning Outcomes
- Students will explain the principles of human-AI interactions.
- Students will describe the advantages and limitations of large-language models for educational assessment.
- Students will compare automation and augmentation in human-AI collaboration.
- Students will iteratively develop prompts and fine-tune pre-trained models for educational assessment tasks.
- Students will identify and mitigate bias in LLM-generated outputs.
- Students will evaluate the reliability, validity, and fairness of AI-assisted assessment tasks.
- Students will design and conduct a research study applying human-AI interaction principles.
Topical Outline
- Introduction to Human-AI interaction
- Transformers and large-language models (LLMs)
- Prompt engineering (zero-shot, few-shot, chain-of-thought, prompt chaining)
- Fine-tuning approaches
- Reliability and consistency of LLM output and human-in-the-loop validation
- Algorithmic fairness
- Automatic item generation, scoring, and feedback
