It is AI’s Turn to Ask Human a Question: Question and Answer Pair Generation for Children Storybooks in FairytaleQA Dataset

Abstract

Existing question answering (QA) techniques are created mainly to answer questions asked by humans. But in educational applications, teachers and parents sometimes may not know what questions they should ask best help children to develop their narrative understanding abilities. We design an automated question-answer generation (QAG) system for education purposes: given a storybook at the kindergarten to eighth-grade level, our system can automatically produce QA pairs that are capable of testing a variety of student comprehension skills. Using a new QA dataset FairytaleQA that has 278 child-friendly storybooks with 10,580 QA pairs labeled by experts, we design a novel QAG system architecture to generate QA pairs. Automatic and human evaluations show that our model outperforms state-of-the-art QAG systems. On top of our QAG system, we also build an interactive story-telling application for future real-world deployment.

Tasks

Reproductions