• DOI:10.1162/tacl_a_00021
Corpus ID: 9192723

Constructing Datasets for Multi-hop Reading Comprehension Across Documents

@article{Welbl2018ConstructingDF,
  title={Constructing Datasets for Multi-hop Reading Comprehension Across Documents},
  author={Johannes Welbl and Pontus Stenetorp and S. Riedel},
  journal={Transactions of the Association for Computational Linguistics},
  year={2018},
  volume={6},
  pages={287-302}
}

Johannes Welbl, Pontus Stenetorp, S. Riedel

Published 2018

Computer Science

Transactions of the Association for Computational Linguistics

Most Reading Comprehension methods limit themselves to queries which can be answered using a single sentence, paragraph, or document. [...] Key Method We devise a methodology to produce datasets for this task, given a collection of query-answer pairs and thematically linked documents. Two datasets from different domains are induced, and we identify potential pitfalls and devise circumvention strategies. We evaluate two previously proposed competitive models and find that one can integrate information across…Expand Abstract

View on MIT Press

mitpressjournals.org

Save to Library

Create Alert

Cite

Launch Research Feed