Unsupervised Text Style Transfer with Padded Masked Language Models
This addresses the problem of text style transfer without parallel data for NLP researchers, offering an incremental improvement over existing methods.
The paper tackles unsupervised text style transfer by proposing Masker, a method that uses masked language models to identify and replace style-specific tokens, achieving competitive performance in unsupervised settings and improving supervised methods by over 10 percentage points in low-resource scenarios.
We propose Masker, an unsupervised text-editing method for style transfer. To tackle cases when no parallel source-target pairs are available, we train masked language models (MLMs) for both the source and the target domain. Then we find the text spans where the two models disagree the most in terms of likelihood. This allows us to identify the source tokens to delete to transform the source text to match the style of the target domain. The deleted tokens are replaced with the target MLM, and by using a padded MLM variant, we avoid having to predetermine the number of inserted tokens. Our experiments on sentence fusion and sentiment transfer demonstrate that Masker performs competitively in a fully unsupervised setting. Moreover, in low-resource settings, it improves supervised methods' accuracy by over 10 percentage points when pre-training them on silver training data generated by Masker.