Tag-Set-Sequence Learning for Generating Question-Answer Pairs

TL;DR

This paper introduces a novel tag-set sequence learning approach for question-answer pair generation that captures syntactic and semantic features, improving over transformer models especially on specific texts.

Contribution

The paper proposes TSS-Learner, a new method leveraging tag-set sequences to enhance question-answer pair generation from sentences, with promising results on small datasets.

Findings

TSS-Learner generates more adequate QAPs than transformer models on certain texts.

Human evaluation shows encouraging quality of generated QAPs.

Method effectively captures syntactic and semantic features for QG.

Abstract

Transformer-based QG models can generate question-answer pairs (QAPs) with high qualities, but may also generate silly questions for certain texts. We present a new method called tag-set sequence learning to tackle this problem, where a tag-set sequence is a sequence of tag sets to capture the syntactic and semantic information of the underlying sentence, and a tag set consists of one or more language feature tags, including, for example, semantic-role-labeling, part-of-speech, named-entity-recognition, and sentiment-indication tags. We construct a system called TSS-Learner to learn tag-set sequences from given declarative sentences and the corresponding interrogative sentences, and derive answers to the latter. We train a TSS-Learner model for the English language using a small training dataset and show that it can indeed generate adequate QAPs for certain texts that transformer-based models do poorly. Human evaluation on the QAPs generated by TSS-Learner over SAT practice reading tests is encouraging.