Bilingual-GAN: Neural Text Generation and Neural Machine Translation as Two Sides of the Same Coin

Ahmad Rashid, Alan Do-Omri, Mehdi Rezagholizadeh, Md. Akmal Haidar, Hamed Sadeghi

Abstract: Latent space based GAN methods and attention based encoder-decoder architectures have achieved impressive results in text generation and Unsupervised NMT respectively. Leveraging the two domains, we propose an adversarial latent space based architecture capable of generating parallel sentences in two languages concurrently and translating bidirectionally. The bilingual generation goal is achieved by sampling from the latent space that is adversarially constrained to be shared between both languages. First an NMT model is trained, with back-translation and an adversarial setup, to enforce a latent state between the two languages. The encoder and decoder are shared for the two translation directions. Next, a GAN is trained to generate ‘synthetic’ code mimicking the languages’ shared latent space. This code is then fed into the decoder to generate text in either language. We perform our experiments on Europarl and Multi30k datasets, on the English-French language pair, and document our performance using both Supervised and Unsupervised NMT.

Keywords: Text Generation, Machine Translation, Deep Learning, GAN

TL;DR: We present a novel method for Bilingual Text Generation producing parallel concurrent sentences in two languages.