Work

Non-Autoregressive Neural Machine Translation: A Call for Clarity

Year: 2022
Type: conference-paper
Abstract: Non-autoregressive approaches aim to improve the inference speed of translation models by only requiring a single forward pass to generate the output sequence instead of iteratively producing each predicted token.Consequently, their translation quality still tends to be inferior to their autoregressive counterparts due to several issues involving output token interdependence.In this work, we take a step back and revisit several techniques that have been proposed for improving non-autoregressive translation models and compare their combined translation quality and speed implications under thirdparty testing environments.We provide novel insights for establishing strong baselines using length prediction or CTC-based architecture variants and contribute standardized BLEU, CHRF++, and TER scores using sacreBLEU on four translation tasks, which crucially have been missing as inconsistencies in the use of tokenized BLEU lead to deviations of up to 1.7 BLEU points.Our open-sourced code is integrated into fairseq for reproducibility. 1
Language: English
Open Access status: hybrid
OpenAlex ID: W4385572929
FWCI: 0.814
Cites: 62
Cited by: 14
Related to: 10