Semi-autoregressive Training Improves Mask-predict Decoding
Marjan Ghazvininejad, Omer Levy, Luke Zettlemoyer . Arxiv 2020 – 53 citations
[Paper]
The recently proposed mask-predict decoding algorithm has narrowed the performance gap between semi-autoregressive machine translation models and the traditional left-to-right approach. We introduce a new training method for conditional masked language models, SMART, which mimics the semi-autoregressive behavior of mask-predict, producing training examples that contain model predictions as part of their inputs. Models trained with SMART produce higher-quality translations when using mask-predict decoding, effectively closing the remaining performance gap with fully autoregressive models.