Guiding Non-Autoregressive Neural Machine Translation Decoding with Reordering Information

Authors

  • Qiu Ran Pattern Recognition Center, WeChat AI, Tencent Inc., China
  • Yankai Lin Pattern Recognition Center, WeChat AI, Tencent Inc., China
  • Peng Li Pattern Recognition Center, WeChat AI, Tencent Inc., China
  • Jie Zhou Pattern Recognition Center, WeChat AI, Tencent Inc., China

Keywords:

Machine Translation & Multilinguality

Abstract

Non-autoregressive neural machine translation (NAT) generates each target word in parallel and has achieved promising inference acceleration. However, existing NAT models still have a big gap in translation quality compared to autoregressive neural machine translation models due to the multimodality problem: the target words may come from multiple feasible translations. To address this problem, we propose a novel NAT framework ReorderNAT which explicitly models the reordering information to guide the decoding of NAT. Specially, ReorderNAT utilizes deterministic and non-deterministic decoding strategies that leverage reordering information as a proxy for the final translation to encourage the decoder to choose words belonging to the same translation. Experimental results on various widely-used datasets show that our proposed model achieves better performance compared to most existing NAT models, and even achieves comparable translation quality as autoregressive translation models with a significant speedup.

Downloads

Published

2021-05-18

How to Cite

Ran, Q., Lin, Y., Li, P., & Zhou, J. (2021). Guiding Non-Autoregressive Neural Machine Translation Decoding with Reordering Information. Proceedings of the AAAI Conference on Artificial Intelligence, 35(15), 13727-13735. Retrieved from https://ojs.aaai.org/index.php/AAAI/article/view/17618

Issue

Section

AAAI Technical Track on Speech and Natural Language Processing II