Multi-Task Learning with Shared Encoder for Non-Autoregressive Machine Translation

Abstract

Non-Autoregressive machine Translation (NAT) models have demonstrated significant inference speedup but suffer from inferior translation accuracy. The common practice to tackle the problem is transferring the Autoregressive machine Translation (AT) knowledge to NAT models, e.g., with knowledge distillation. In this work, we hypothesize and empirically verify that AT and NAT encoders capture different linguistic properties of source sentences. Therefore, we propose to adopt Multi-Task learning to transfer the AT knowledge to NAT models through encoder sharing. Specifically, we take the AT model as an auxiliary task to enhance NAT model performance. Experimental results on WMT14 English-German and WMT16 English-Romanian datasets show that the proposed Multi-Task NAT achieves significant improvements over the baseline NAT models. Furthermore, the performance on large-scale WMT19 and WMT20 English-German datasets confirm the consistency of our proposed method. In addition, experimental results demonstrate that our Multi-Task NAT is complementary to knowledge distillation, the standard knowledge transfer method for NAT.

NAACL 2021

References in corpus (11)

• Adam: A Method for Stochastic Optimization
• Neural Machine Translation by Jointly Learning to Align and Translate
• An Overview of Multi-Task Learning in Deep Neural Networks
• Non-Autoregressive Neural Machine Translation
• SentEval: An Evaluation Toolkit for Universal Sentence Representations
• What you can cram into a single vector: Probing sentence embeddings for linguistic properties
• Levenshtein Transformer
• Understanding and Improving Lexical Choice in Non-Autoregressive Translation
• Understanding Learning Dynamics Of Language Models with SVCCA
• compare-mt: A Tool for Holistic Comparison of Language Generation Systems
• Data Rejuvenation: Exploiting Inactive Training Examples for Neural Machine Translation