Improvement in Machine Translation with Generative Adversarial Networks
This is an incremental improvement for machine translation systems, potentially benefiting users of translation tools.
The paper tackles the problem of improving machine translation by using Generative Adversarial Networks (GANs) to transform non-fluent English sentences into fluent ones, trained only on monolingual corpora, with results showing improvement over phrase-based machine translation in some cases.
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 $λ$ 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.