WMT14 En-De Аударма Тапсырмасы

AbstractNon-autoregressive (NAR) neural machine translation is usually done via knowledge distillation from an autoregressive (AR) model. Under this framework, we leverage large monolingual corpora to improve the NAR model’s performance, …...

AbstractNon-autoregressive translation (NAT) predicts all the target tokens in parallel and significantly speeds up the inference process. The Conditional Masked Language Model (CMLM) is a strong baseline of NAT. It …...