Question Generation for Assessing Early Literacy Reading Comprehension

TL;DR

This paper introduces a new method for generating comprehension questions tailored to early learners, ensuring comprehensive coverage, adaptability to individual skills, and diverse question types to improve reading assessment.

Contribution

It presents a novel question generation approach that adapts to learner proficiency and covers material thoroughly, advancing AI-driven literacy assessment tools.

Findings

Effective question diversity across difficulty levels

High coverage of underlying reading material

Potential for autonomous AI literacy instruction

Abstract

Assessment of reading comprehension through content-based interactions plays an important role in the reading acquisition process. In this paper, we propose a novel approach for generating comprehension questions geared to K-2 English learners. Our method ensures complete coverage of the underlying material and adaptation to the learner's specific proficiencies, and can generate a large diversity of question types at various difficulty levels to ensure a thorough evaluation. We evaluate the performance of various language models in this framework using the FairytaleQA dataset as the source material. Eventually, the proposed approach has the potential to become an important part of autonomous AI-driven English instructors.