So Different Yet So Alike! Constrained Unsupervised Text Style Transfer

TL;DR

This paper presents a novel constrained unsupervised text style transfer method using cooperative losses in GANs to better preserve attributes like length and descriptiveness across domains.

Contribution

It introduces cooperative contrastive and classification losses in GANs to explicitly maintain multiple text attributes during style transfer, improving quality and constraint preservation.

Findings

Enhanced attribute preservation across multiple datasets

Improved text quality according to automated and human evaluations

Effective handling of multiple attribute changes in style transfer

Abstract

Automatic transfer of text between domains has become popular in recent times. One of its aims is to preserve the semantic content of text being translated from source to target domain. However, it does not explicitly maintain other attributes between the source and translated text, for e.g., text length and descriptiveness. Maintaining constraints in transfer has several downstream applications, including data augmentation and de-biasing. We introduce a method for such constrained unsupervised text style transfer by introducing two complementary losses to the generative adversarial network (GAN) family of models. Unlike the competing losses used in GANs, we introduce cooperative losses where the discriminator and the generator cooperate and reduce the same loss. The first is a contrastive loss and the second is a classification loss, aiming to regularize the latent space further and bring similar sentences across domains closer together. We demonstrate that such training retains lexical, syntactic, and domain-specific constraints between domains for multiple benchmark datasets, including ones where more than one attribute change. We show that the complementary cooperative losses improve text quality, according to both automated and human evaluation measures.