Lesson Plan Generation Using Natural Language Processing: Prompting Best Practices With OpenAI's GPT-3 Model (Poster 27)

• In Event: AERA Poster Session 10
  In Poster Session: Advanced Technologies for Learning SIG Poster Session: Data Science and Artificial Intelligence in Education (San Diego)

Abstract

In the field of Natural Language Processing ( NLP), advances in computing power have allowed researchers to create Large Language Models (LLMs) capable of generating large bodies of text. LLMs are of particular interest for their ability to mimic linguistic structures and other forms of human communication with very few examples or prompting, sometimes as little as zero examples within a one sentence prompt. One particular model, OpenAI’s GPT-3, has captured the public imagination with its ability to generate natural-sounding, structured language like poems, stories, or songs without seeing examples. So what happens when models like GPT-3 are asked to generate K-12 lesson plans? This paper examines these results, and explores prompting best practices.

Author

• Joel Anderson Walsh, Privateer Space