Latent Code and Text-based Generative Adversarial Networks for Soft-text Generation
This work addresses text generation for NLP applications, but it appears incremental as it builds on existing GAN and autoencoder frameworks.
The authors tackled the problem of text generation with GANs by introducing Soft-GAN and LATEXT-GAN, which use continuous sentence representations and hybrid discriminations, and showed that these techniques outperform traditional GAN-based methods on SNLI and Image COCO datasets.
Text generation with generative adversarial networks (GANs) can be divided into the text-based and code-based categories according to the type of signals used for discrimination. In this work, we introduce a novel text-based approach called Soft-GAN to effectively exploit GAN setup for text generation. We demonstrate how autoencoders (AEs) can be used for providing a continuous representation of sentences, which we will refer to as soft-text. This soft representation will be used in GAN discrimination to synthesize similar soft-texts. We also propose hybrid latent code and text-based GAN (LATEXT-GAN) approaches with one or more discriminators, in which a combination of the latent code and the soft-text is used for GAN discriminations. We perform a number of subjective and objective experiments on two well-known datasets (SNLI and Image COCO) to validate our techniques. We discuss the results using several evaluation metrics and show that the proposed techniques outperform the traditional GAN-based text-generation methods.