Abstract

Current Machine Translation (MT) models still struggle with more challenging input, such as noisy data and tail-end words and phrases. Several works have addressed this robustness issue by identifying specific categories of noise and variation then tuning models to perform better on them. An important yet under-studied category involves minor variations in nuance (non-typos) that preserve meaning w.r.t. the target language. We introduce and formalize this category as Natural Asemantic Variation (NAV) and investigate it in the context of MT robustness. We find that existing MT models fail when presented with NAV data, but we demonstrate strategies to improve performance on NAV by fine-tuning them with human-generated variations. We also show that NAV robustness can be transferred across languages and find that synthetic perturbations can achieve some but not all of the benefits of organic NAV data.

Metadata

publication

arXiv preprint arXiv:2205.12514, 2022

year

2022

publication date

2022/5/25

authors

Jacob Bremerman, Xiang Ren, Jonathan May

link

https://arxiv.org/abs/2205.12514

resource_link

https://arxiv.org/pdf/2205.12514

journal

arXiv preprint arXiv:2205.12514