@inproceedings{kojima-etal-2021-making,
    title = "Making Use of Latent Space in Language {GAN}s for Generating Diverse Text without Pre-training",
    author = "Kojima, Takeshi  and
      Iwasawa, Yusuke  and
      Matsuo, Yutaka",
    editor = "Sorodoc, Ionut-Teodor  and
      Sushil, Madhumita  and
      Takmaz, Ece  and
      Agirre, Eneko",
    booktitle = "Proceedings of the 16th Conference of the European Chapter of the Association for Computational Linguistics: Student Research Workshop",
    month = apr,
    year = "2021",
    address = "Online",
    publisher = "Association for Computational Linguistics",
    url = "https://preview.aclanthology.org/Add-Cong-Liu-Florida-Atlantic-University-author-id/2021.eacl-srw.23/",
    doi = "10.18653/v1/2021.eacl-srw.23",
    pages = "175--182",
    abstract = "Generating diverse texts is an important factor for unsupervised text generation. One approach is to produce the diversity of texts conditioned by the sampled latent code. Although several generative adversarial networks (GANs) have been proposed thus far, these models still suffer from mode-collapsing if the models are not pre-trained. In this paper, we propose a GAN model that aims to improve the approach to generating diverse texts conditioned by the latent space. The generator of our model uses Gumbel-Softmax distribution for the word sampling process. To ensure that the text is generated conditioned upon the sampled latent code, reconstruction loss is introduced in our objective function. The discriminator of our model iteratively inspects incomplete partial texts and learns to distinguish whether they are real or fake by using the standard GAN objective function. Experimental results using the COCO Image Captions dataset show that, although our model is not pre-trained, the performance of our model is quite competitive with the existing baseline models, which requires pre-training."
}