Question-type Driven Question Generation

TL;DR

This paper introduces a method that predicts question types from answers and context, then incorporates this into a seq2seq model to improve question generation accuracy, achieving state-of-the-art results on SQuAD and MARCO datasets.

Contribution

It presents a novel approach that predicts question types and integrates them into question generation, addressing the mismatch issue in prior methods.

Findings

Significant improvement in question type prediction accuracy.

State-of-the-art question generation results on SQuAD.

Enhanced alignment between question types and answers.

Abstract

Question generation is a challenging task which aims to ask a question based on an answer and relevant context. The existing works suffer from the mismatching between question type and answer, i.e. generating a question with type $how$ while the answer is a personal name. We propose to automatically predict the question type based on the input answer and context. Then, the question type is fused into a seq2seq model to guide the question generation, so as to deal with the mismatching problem. We achieve significant improvement on the accuracy of question type prediction and finally obtain state-of-the-art results for question generation on both SQuAD and MARCO datasets.