Abstract
With the advent of the Transformer architecture, Neural Machine Translation (NMT) results have shown great improvement lately. However, results in low-resource conditions still lag behind in both bilingual and multilingual setups, due to the limited amount of available monolingual and/or parallel data; hence, the need for methods addressing data scarcity in an efficient, and explainable way, is eminent. We propose an explainability-based training approach for NMT, applied in Unsupervised and Supervised model training, for translation of three languages of varying resources, French, Gujarati, Kazakh, to and from English. Our results show our method can be promising, particularly when training in low-resource conditions, outperforming simple training baselines; though the improvement is marginal, it sets the ground for further exploration of the approach and the parameters, and its extension to other languages.
Abstract (translated)
随着Transformer架构的出现,神经机器翻译(NMT)的结果最近取得了很大的改善。然而,在双语和多语言设置中,低资源条件下的结果仍然落后,这是因为可用的小量单语和/或并行数据有限;因此,在高效且可解释的条件下解决数据稀缺问题,具有重要的意义。我们提出了一个以解释性为基础的NMT训练方法,应用于无监督和监督模型训练,用于三种资源丰富程度不同的语言(法语、印地语、哈萨克语)到英语的翻译。我们的结果表明,我们的方法具有很好的前景,尤其是在低资源条件下,能够超越简单的训练基线;尽管改进非常微小,但它为该方法及其在其他国家语言上的扩展奠定了基础。
URL
https://arxiv.org/abs/2312.00214