Bi-SimCut: A Simple Strategy for Boosting Neural Machine Translation
Preprint
Abstract
We introduce Bi-SimCut: a simple but effective training strategy to boost neural machine translation (NMT) performance. It consists of two procedures: bidirectional pretraining and unidirectional finetuning. Both procedures utilize SimCut, a simple regularization method that forces the consistency between the output distributions of the original and the cutoff sentence pairs. Without leveraging extra dataset via back-translation or integrating large-scale pretrained model, Bi-SimCut achieves strong translation performance across five translation benchmarks (data sizes range from 160K to 20.2M): BLEU scores of 31.16 for en -> de and 38.37 for de -> en on the IWSLT14 dataset, 30.78 for en -> de and 35.15 for de -> en on the WMT14 dataset, and 27.17 for zh -> en on the WMT17 dataset. SimCut is not a new method, but a version of Cutoff (Shen et al., 2020) simplified and adapted for NMT, and it could be considered as a perturbation-based method. Given the universality and simplicity of SimCut and Bi-SimCut, we believe they can serve as strong baselines for future NMT research.Keywords
All Related Versions
This publication has 16 references indexed in Scilit:
- BERT, mBERT, or BiBERT? A Study on Contextualized Embeddings for Neural Machine Translation2021
- Improving Neural Machine Translation by Bidirectional Training2021
- Manifold Adversarial Augmentation for Neural Machine Translation2021
- Multilingual Translation from Denoising Pre-Training2021
- UniDrop: A Simple yet Effective Technique to Improve Transformer without Extra Cost2021
- mT5: A Massively Multilingual Pre-trained Text-to-Text Transformer2021
- Rethinking Perturbations in Encoder-Decoders for Fast Training2021
- Multilingual Denoising Pre-training for Neural Machine TranslationTransactions of the Association for Computational Linguistics, 2020
- Leveraging Pre-trained Checkpoints for Sequence Generation TasksTransactions of the Association for Computational Linguistics, 2020
- SentencePiece: A simple and language independent subword tokenizer and detokenizer for Neural Text Processing2018