Improvement in Machine Translation with Generative Adversarial Networks

TL;DR

This paper presents a GAN-based approach to improve machine translation by transforming non-fluent sentences into fluent ones using monolingual data, showing promising results with transformer generators.

Contribution

Introduces a novel GAN architecture inspired by RelGAN and NMT-GAN for monolingual sentence fluency enhancement in machine translation.

Findings

GAN with transformer generator outperforms some phrase-based methods

Model effectively learns to improve sentence fluency from monolingual data

Provides a proof-of-concept for GAN-based translation enhancement

Abstract

In this paper, we explore machine translation improvement via Generative Adversarial Network (GAN) architecture. We take inspiration from RelGAN, a model for text generation, and NMT-GAN, an adversarial machine translation model, to implement a model that learns to transform awkward, non-fluent English sentences to fluent ones, while only being trained on monolingual corpora. We utilize a parameter $\lambda$ to control the amount of deviation from the input sentence, i.e. a trade-off between keeping the original tokens and modifying it to be more fluent. Our results improved upon phrase-based machine translation in some cases. Especially, GAN with a transformer generator shows some promising results. We suggests some directions for future works to build upon this proof-of-concept.