Document-level neural machine translation (DocNMT) aims to generate translations that are both coherent and cohesive, in contrast to its sentence-level counterpart. However, due to its longer input length and limited availability of training data, DocNMT often faces the challenge of data sparsity. To overcome this issue, we propose a novel Importance-Aware Data Augmentation (IADA) algorithm for DocNMT that augments the training data based on token importance information estimated by the norm of hidden states and training gradients. We conduct comprehensive experiments on three widely-used DocNMT benchmarks. Our empirical results show that our proposed IADA outperforms strong DocNMT baselines as well as several data augmentation approaches, with statistical significance on both sentence-level and document-level BLEU.
@article{arxiv.2401.15360,
title = {Importance-Aware Data Augmentation for Document-Level Neural Machine Translation},
author = {Minghao Wu and Yufei Wang and George Foster and Lizhen Qu and Gholamreza Haffari},
journal= {arXiv preprint arXiv:2401.15360},
year = {2024}
}
Comments
13 pages, 4 figures, 7 tables, accepted by EACL2024 main conference