Multilingual Neural Machine Translation with Deep Encoder and Multiple Shallow Decoders

Xiang Kong, Adithya Renduchintala, James Cross, Yuqing Tang, Jiatao Gu, Xian Li


Abstract
Recent work in multilingual translation advances translation quality surpassing bilingual baselines using deep transformer models with increased capacity. However, the extra latency and memory costs introduced by this approach may make it unacceptable for efficiency-constrained applications. It has recently been shown for bilingual translation that using a deep encoder and shallow decoder (DESD) can reduce inference latency while maintaining translation quality, so we study similar speed-accuracy trade-offs for multilingual translation. We find that for many-to-one translation we can indeed increase decoder speed without sacrificing quality using this approach, but for one-to-many translation, shallow decoders cause a clear quality drop. To ameliorate this drop, we propose a deep encoder with multiple shallow decoders (DEMSD) where each shallow decoder is responsible for a disjoint subset of target languages. Specifically, the DEMSD model with 2-layer decoders is able to obtain a 1.8x speedup on average compared to a standard transformer model with no drop in translation quality.
Anthology ID:
2021.eacl-main.138
Volume:
Proceedings of the 16th Conference of the European Chapter of the Association for Computational Linguistics: Main Volume
Month:
April
Year:
2021
Address:
Online
Venue:
EACL
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
1613–1624
Language:
URL:
https://aclanthology.org/2021.eacl-main.138
DOI:
10.18653/v1/2021.eacl-main.138
Bibkey:
Cite (ACL):
Xiang Kong, Adithya Renduchintala, James Cross, Yuqing Tang, Jiatao Gu, and Xian Li. 2021. Multilingual Neural Machine Translation with Deep Encoder and Multiple Shallow Decoders. In Proceedings of the 16th Conference of the European Chapter of the Association for Computational Linguistics: Main Volume, pages 1613–1624, Online. Association for Computational Linguistics.
Cite (Informal):
Multilingual Neural Machine Translation with Deep Encoder and Multiple Shallow Decoders (Kong et al., EACL 2021)
Copy Citation:
PDF:
https://aclanthology.org/2021.eacl-main.138.pdf