Viterbi Decoding of Directed Acyclic Transformer for Non-Autoregressive Machine Translation

Chenze Shao, Zhengrui Ma, Yang Feng


Abstract
Non-autoregressive models achieve significant decoding speedup in neural machine translation but lack the ability to capture sequential dependency. Directed Acyclic Transformer (DA-Transformer) was recently proposed to model sequential dependency with a directed acyclic graph. Consequently, it has to apply a sequential decision process at inference time, which harms the global translation accuracy. In this paper, we present a Viterbi decoding framework for DA-Transformer, which guarantees to find the joint optimal solution for the translation and decoding path under any length constraint. Experimental results demonstrate that our approach consistently improves the performance of DA-Transformer while maintaining a similar decoding speedup.
Anthology ID:
2022.findings-emnlp.322
Volume:
Findings of the Association for Computational Linguistics: EMNLP 2022
Month:
December
Year:
2022
Address:
Abu Dhabi, United Arab Emirates
Editors:
Yoav Goldberg, Zornitsa Kozareva, Yue Zhang
Venue:
Findings
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
4390–4397
Language:
URL:
https://aclanthology.org/2022.findings-emnlp.322
DOI:
10.18653/v1/2022.findings-emnlp.322
Bibkey:
Cite (ACL):
Chenze Shao, Zhengrui Ma, and Yang Feng. 2022. Viterbi Decoding of Directed Acyclic Transformer for Non-Autoregressive Machine Translation. In Findings of the Association for Computational Linguistics: EMNLP 2022, pages 4390–4397, Abu Dhabi, United Arab Emirates. Association for Computational Linguistics.
Cite (Informal):
Viterbi Decoding of Directed Acyclic Transformer for Non-Autoregressive Machine Translation (Shao et al., Findings 2022)
Copy Citation:
PDF:
https://aclanthology.org/2022.findings-emnlp.322.pdf