Adaptive Sparse Transformer for Multilingual Translation

by   Hongyu Gong, et al.

Multilingual machine translation has attracted much attention recently due to its support of knowledge transfer among languages and the low cost of training and deployment compared with numerous bilingual models. A known challenge of multilingual models is the negative language interference. In order to enhance the translation quality, deeper and wider architectures are applied to multilingual modeling for larger model capacity, which suffers from the increased inference cost at the same time. It has been pointed out in recent studies that parameters shared among languages are the cause of interference while they may also enable positive transfer. Based on these insights, we propose an adaptive and sparse architecture for multilingual modeling, and train the model to learn shared and language-specific parameters to improve the positive transfer and mitigate the interference. The sparse architecture only activates a subnetwork which preserves inference efficiency, and the adaptive design selects different subnetworks based on the input languages. Evaluated on multilingual translation across multiple public datasets, our model outperforms strong baselines in terms of translation quality without increasing the inference cost.


page 1

page 2

page 3

page 4


HLT-MT: High-resource Language-specific Training for Multilingual Neural Machine Translation

Multilingual neural machine translation (MNMT) trained in multiple langu...

Pay Better Attention to Attention: Head Selection in Multilingual and Multi-Domain Sequence Modeling

Multi-head attention has each of the attention heads collect salient inf...

Tricks for Training Sparse Translation Models

Multi-task learning with an unbalanced data distribution skews model lea...

Multilingual Machine Translation with Hyper-Adapters

Multilingual machine translation suffers from negative interference acro...

On Negative Interference in Multilingual Models: Findings and A Meta-Learning Treatment

Modern multilingual models are trained on concatenated text from multipl...

Can Multilinguality benefit Non-autoregressive Machine Translation?

Non-autoregressive (NAR) machine translation has recently achieved signi...

Alternative Input Signals Ease Transfer in Multilingual Machine Translation

Recent work in multilingual machine translation (MMT) has focused on the...