arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 1907.00409cs.CL
收藏

面向低资源语言的语言模型微调技术评估

Evaluating Language Model Finetuning Techniques for Low-resource Languages

Center for Language Technologies(语言技术中心);College of Computer Studies(计算机研究学院);De La Salle University(德拉萨大学)

Jan Christian Blaise Cruz, Charibeth Cheng

摘要:

与主流语言(如英语和法语)不同,低资源语言通常缺乏专家注释的语料库和基准资源,因此很难直接应用最先进的技术。在本文中,我们通过两种方式缓解资源匮乏的菲律宾语的稀缺问题。首先,我们介绍了一个新的基准语言建模数据集在菲律宾,我们称之为WikiText-TL-39。其次,我们表明,语言模型微调技术,如BERT和ULMFiT可用于一致的训练强大的分类器在低资源设置,经历了最多0.0782增加验证错误时,训练样本的数量从10 K减少到1 K,而微调使用私人持有的情感数据集。

英文摘要:

Unlike mainstream languages (such as English and French), low-resource languages often suffer from a lack of expert-annotated corpora and benchmark resources that make it hard to apply state-of-the-art techniques directly. In this paper, we alleviate this scarcity problem for the low-resourced Filipino language in two ways. First, we introduce a new benchmark language modeling dataset in Filipino which we call WikiText-TL-39. Second, we show that language model finetuning techniques such as BERT and ULMFiT can be used to consistently train robust classifiers in low-resource settings, experiencing at most a 0.0782 increase in validation error when the number of training examples is decreased from 10K to 1K while finetuning using a privately-held sentiment dataset.

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。

↑