We study the task of generating from Wikipedia articles question-answer pairs that cover content beyond a single sentence. We propose a neural network approach that incorporates coreference knowledge via a novel gating mechanism. Compared to models that only take into account sentence-level information (Heilman and Smith, 2010; Du et al., 2017; Zhou et al., 2017), we find that the linguistic knowledge introduced by the coreference representation aids question generation significantly, producing models that outperform the current state-of-the-art. We apply our system (composed of an answer span extraction system and the passage-level QG system) to the 10,000 top-ranking Wikipedia articles and create a corpus of over one million question-answer pairs. We also provide a qualitative analysis for this large-scale generated corpus from Wikipedia.
@article{arxiv.1805.05942,
title = {Harvesting Paragraph-Level Question-Answer Pairs from Wikipedia},
author = {Xinya Du and Claire Cardie},
journal= {arXiv preprint arXiv:1805.05942},
year = {2018}
}