论文标题
非自动回归神经机器翻译：呼吁清晰
Non-Autoregressive Neural Machine Translation: A Call for Clarity
论文作者
论文摘要
非自动进取的方法旨在通过仅需要单个前向通过即可生成输出顺序，而不是迭代产生每个预测的令牌，旨在提高翻译模型的推理速度。因此，由于几个涉及输出令牌相互依存的问题，它们的翻译质量仍然往往不如其自回归同行。在这项工作中，我们退后一步，重新审视了一些用于改善非自动回旋翻译模型并比较其在第三方测试环境下的合并翻译质量和速度含义的技术。我们提供了使用长度预测或基于CTC的体系结构变体建立强基础的新颖见解，并在四个翻译任务上使用Sacrebleu贡献了标准化的BLEU，CHRF ++和TER分数，这在四个翻译任务上一直缺失，因为这些任务是不一致的，在使用标记的BLEU偏离偏离偏差时，它最多可达1.7 bleuupoint。我们的开源代码已集成到Fairseq，以供可重复使用。
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 third-party 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.
