Incorporating Discriminator in Sentence Generation: a Gibbs Sampling Method

Authors

  • Jinyue Su Fudan University
  • Jiacheng Xu Fudan University
  • Xipeng Qiu Fudan University
  • Xuanjing Huang Fudan University

Keywords:

natural language generation, MCMC, gibbs sampling

Abstract

Generating plausible and fluent sentence with desired properties has long been a challenge. Most of the recent works use recurrent neural networks (RNNs) and their variants to predict following words given previous sequence and target label. In this paper, we propose a novel framework to generate constrained sentences via Gibbs Sampling. The candidate sentences are revised and updated iteratively, with sampled new words replacing old ones. Our experiments show the effectiveness of the proposed method to generate plausible and diverse sentences.

Downloads

Published

2018-04-27

How to Cite

Su, J., Xu, J., Qiu, X., & Huang, X. (2018). Incorporating Discriminator in Sentence Generation: a Gibbs Sampling Method. Proceedings of the AAAI Conference on Artificial Intelligence, 32(1). Retrieved from https://ojs.aaai.org/index.php/AAAI/article/view/11990