gigaword

  • Description:

Headline-generation on a corpus of article pairs from Gigaword consisting of around 4 million articles. Use the 'org_data' provided by https://github.com/microsoft/unilm/ which is identical to https://github.com/harvardnlp/sent-summary but with better format.

There are two features: - document: article. - summary: headline.

Split Examples
'test' 1,951
'train' 3,803,957
'validation' 189,651
  • Feature structure:
FeaturesDict({
    'document': Text(shape=(), dtype=string),
    'summary': Text(shape=(), dtype=string),
})
  • Feature documentation:
Feature Class Shape Dtype Description
FeaturesDict
document Text string
summary Text string
  • Citation:
@article{graff2003english,
  title={English gigaword},
  author={Graff, David and Kong, Junbo and Chen, Ke and Maeda, Kazuaki},
  journal={Linguistic Data Consortium, Philadelphia},
  volume={4},
  number={1},
  pages={34},
  year={2003}
}

@article{Rush_2015,
   title={A Neural Attention Model for Abstractive Sentence Summarization},
   url={http://dx.doi.org/10.18653/v1/D15-1044},
   DOI={10.18653/v1/d15-1044},
   journal={Proceedings of the 2015 Conference on Empirical Methods in Natural Language Processing},
   publisher={Association for Computational Linguistics},
   author={Rush, Alexander M. and Chopra, Sumit and Weston, Jason},
   year={2015}
}