@InProceedings{gehrmann-EtAl:2018:W18-65,
  author    = {Gehrmann, Sebastian  and  Dai, Falcon  and  Elder, Henry  and  Rush, Alexander},
  title     = {End-to-End Content and Plan Selection for Data-to-Text Generation},
  booktitle = {Proceedings of the 11th International Conference on Natural Language Generation},
  month     = {September},
  year      = {2018},
  address   = {Tilburg University, The Netherlands},
  publisher = {Association for Computational Linguistics},
  pages     = {46--56},
  abstract  = {Learning to generate fluent natural language from structured data with neural networks has become an common approach for NLG. This problem can be challenging when the form of the structured data varies between examples. This paper presents a survey of several extensions to sequence-to-sequence models to account for the latent content selection process, particularly variants of copy attention and coverage decoding. We further propose a training method based on diverse ensembling to encourage the model to learn latent generation of plans during training. An empirical evaluation of these techniques shows an increase in quality of generated text across five automated metrics, as well as human evaluation.},
  url       = {http://www.aclweb.org/anthology/W18-6505}
}