Self-Training Sampling with Monolingual Data Uncertainty for Neural Machine Translation
arXiv:2106.00941
Abstract
Self-training has proven effective for improving NMT performance by augmenting model training with synthetic parallel data. The common practice is to construct synthetic data based on a randomly sampled subset of large-scale monolingual data, which we empirically show is sub-optimal. In this work, we propose to improve the sampling procedure by selecting the most informative monolingual sentences to complement the parallel data. To this end, we compute the uncertainty of monolingual sentences using the bilingual dictionary extracted from the parallel data. Intuitively, monolingual sentences with lower uncertainty generally correspond to easy-to-translate patterns which may not provide additional gains. Accordingly, we design an uncertainty-based sampling strategy to efficiently exploit the monolingual data for self-training, in which monolingual sentences with higher uncertainty would be sampled with higher probability. Experimental results on large-scale WMT EnglishGerman and EnglishChinese datasets demonstrate the effectiveness of the proposed approach. Extensive analyses suggest that emphasizing the learning on uncertain monolingual sentences by our approach does improve the translation quality of high-uncertainty sentences and also benefits the prediction of low-frequency words at the target side.
ACL 2021 main conference, long paper, 11 pages
References in corpus (7)
- Multilingual Denoising Pre-training for Neural Machine Translation
- Pay Less Attention with Lightweight and Dynamic Convolutions
- A Simple, Fast Diverse Decoding Algorithm for Neural Generation
- Translationese in Machine Translation Evaluation
- TranSmart: A Practical Interactive Machine Translation System
- Data Rejuvenation: Exploiting Inactive Training Examples for Neural Machine Translation
- Exploiting Unsupervised Data for Emotion Recognition in Conversations