English

Scalable Transformers for Neural Machine Translation

Computation and Language 2021-06-21 v2

Abstract

Transformer has been widely adopted in Neural Machine Translation (NMT) because of its large capacity and parallel training of sequence generation. However, the deployment of Transformer is challenging because different scenarios require models of different complexities and scales. Naively training multiple Transformers is redundant in terms of both computation and memory. In this paper, we propose a novel Scalable Transformers, which naturally contains sub-Transformers of different scales and have shared parameters. Each sub-Transformer can be easily obtained by cropping the parameters of the largest Transformer. A three-stage training scheme is proposed to tackle the difficulty of training the Scalable Transformers, which introduces additional supervisions from word-level and sequence-level self-distillation. Extensive experiments were conducted on WMT EN-De and En-Fr to validate our proposed Scalable Transformers.

Keywords

Cite

@article{arxiv.2106.02242,
  title  = {Scalable Transformers for Neural Machine Translation},
  author = {Peng Gao and Shijie Geng and Yu Qiao and Xiaogang Wang and Jifeng Dai and Hongsheng Li},
  journal= {arXiv preprint arXiv:2106.02242},
  year   = {2021}
}

Comments

Mostly overlapping with version 1, with minor updates/revisions

R2 v1 2026-06-24T02:49:29.152Z