Using OpenAI GPT to Generate Reading Comprehension Items

• In Event: Applications of AI

Abstract

The purpose of this study is to introduce and evaluate a method for generating reading comprehension items using
automatic item generation. The model generates reading comprehension items where the student is required to identify
one irrelevant sentence in passage that contains 5 sentences. The passages were generated using OpenAI GPT-3.

Authors

• Ayfer Sayin, Gazi University

• Mark J Gierl, University of Alberta