Towards more variation in text generation:
Developing and evaluating variation models for choice of referential form

This page aims to introduce additional materials for this article, published in the Proceedings of the 54th Annual Meeting of the Association for Computational Linguistics (ACL 2016).

Abstract

In this study, we introduce a non-deterministic method for referring expression generation. We describe two models that account for individual variation in the choice of referential form in automatically generated text: a Naive Bayes model and a Recurrent Neural Network. Both are evaluated using the VaREG corpus. Then we select the best performing model to generate referential forms in texts from the GREC-2.0 corpus and conduct an evaluation experiment in which humans judge the coherence and comprehensibility of the generated texts, comparing them both with the original references and those produced by a random baseline model.

Corpora

VaREG Corpus Click here to have more information about the VaREG corpus.

Re-annotated GREC-2.0 corpus Click here to download the re-annotated version of the GREC-2.0 corpus, introduced in the study.

Human Evaluation

This study introduces a human evaluation, performed to evaluate the coherence and comprehensibility of a set of texts. The experiment can be accessed here, and the stimuli can be downloaded here

Presentation

The slides for the talk at ACL 2016 can be found here.

Citation

@InProceedings{castroferreira-krahmer-wubben:2016:P16-1,
  author    = {Castro Ferreira, Thiago  and  Krahmer, Emiel  and  Wubben, Sander},
  title     = {Towards more variation in text generation: Developing and evaluating variation models for choice of referential form},
  booktitle = {Proceedings of the 54th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)},
  month     = {August},
  year      = {2016},
  address   = {Berlin, Germany},
  publisher = {Association for Computational Linguistics},
  pages     = {568--577},
  url       = {http://www.aclweb.org/anthology/P16-1054}
}