Towards Universality in Multilingual Text Rewriting

TL;DR

This paper introduces a universal multilingual text rewriter capable of handling multiple languages and attributes, achieving state-of-the-art results in unsupervised translation and zero-shot attribute transfer.

Contribution

The work presents a novel universal model that rewrites text across languages and attributes without supervision, enabling zero-shot sentiment and formality-sensitive translation.

Findings

Achieved state-of-the-art unsupervised translation results.

Demonstrated zero-shot sentiment transfer in non-English languages.

Enabled simultaneous modification of multiple attributes.

Abstract

In this work, we take the first steps towards building a universal rewriter: a model capable of rewriting text in any language to exhibit a wide variety of attributes, including styles and languages, while preserving as much of the original semantics as possible. In addition to obtaining state-of-the-art results on unsupervised translation, we also demonstrate the ability to do zero-shot sentiment transfer in non-English languages using only English exemplars for sentiment. We then show that our model is able to modify multiple attributes at once, for example adjusting both language and sentiment jointly. Finally, we show that our model is capable of performing zero-shot formality-sensitive translation.