arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2004.07324cs.CL
收藏

基于知识蒸馏的多领域神经机器翻译模型的构建

Building a Multi-domain Neural Machine Translation Model using Knowledge Distillation

BNP Paribas(法国巴黎银行)

Idriss Mghabbar, Pirashanth Ratnamogan

摘要:

缺乏专业数据使得构建多领域神经机器翻译工具具有挑战性。尽管处理低资源语言的新兴文献开始显示出有希望的结果,但大多数最先进的模型使用了数百万个句子。如今,大多数多领域自适应技术是基于复杂且精密的架构,其不适用于现实世界的应用。到目前为止,没有一种可伸缩的方法比简单而有效的混合微调更好,即用所有专用数据和通用数据的混合来微调通用模型。在本文中,我们提出了一个新的训练管道,其中知识蒸馏和多个专业教师允许我们有效地微调模型,而不会在推理时增加新的成本。我们的实验表明,我们的训练管道可以在BLEU中将多域翻译的性能提高到2个点,而不是在具有2个,3个和4个域的配置中进行微调。

英文摘要:

Lack of specialized data makes building a multi-domain neural machine translation tool challenging. Although emerging literature dealing with low resource languages starts to show promising results, most state-of-the-art models used millions of sentences. Today, the majority of multi-domain adaptation techniques are based on complex and sophisticated architectures that are not adapted for real-world applications. So far, no scalable method is performing better than the simple yet effective mixed-finetuning, i.e finetuning a generic model with a mix of all specialized data and generic data. In this paper, we propose a new training pipeline where knowledge distillation and multiple specialized teachers allow us to efficiently finetune a model without adding new costs at inference time. Our experiments demonstrated that our training pipeline allows improving the performance of multi-domain translation over finetuning in configurations with 2, 3, and 4 domains by up to 2 points in BLEU.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑