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 2010.01764cs.LGcs.AIstat.ML
收藏

预训练Transformers的任务无关数据增强有多有效?

How Effective is Task-Agnostic Data Augmentation for Pretrained Transformers?

Apple Inc.(苹果公司)

Shayne Longpre, Yu Wang, Christopher DuBois

摘要:

任务无关形式的数据增强已被证明在计算机视觉中广泛有效,即使在预训练模型上也是如此。在NLP中,类似的结果最常见于低数据状态,非预训练模型或预训练模型的情况。在本文中,我们问如何有效的这些技术真的是当应用到预先训练的Transformers。使用两种流行的任务无关数据增强(不针对任何特定任务),简单的数据增强(Wei和Zou,2019)和反向翻译(Sennrichet等人,2015),我们对它们在5个分类任务、6个数据集和3个现代预训练Transformers变体(包括BERT、XLNet和RoBERTa)中的效果进行了系统的检查。我们观察到一个负面的结果,发现以前报告的非预训练模型的强大改进的技术未能持续提高预训练Transformers的性能,即使训练数据是有限的。我们希望这种实证分析有助于告知从业者,数据增强技术可能会带来改进。

英文摘要:

Task-agnostic forms of data augmentation have proven widely effective in computer vision, even on pretrained models. In NLP similar results are reported most commonly for low data regimes, non-pretrained models, or situationally for pretrained models. In this paper we ask how effective these techniques really are when applied to pretrained transformers. Using two popular varieties of task-agnostic data augmentation (not tailored to any particular task), Easy Data Augmentation (Wei and Zou, 2019) and Back-Translation (Sennrichet al., 2015), we conduct a systematic examination of their effects across 5 classification tasks, 6 datasets, and 3 variants of modern pretrained transformers, including BERT, XLNet, and RoBERTa. We observe a negative result, finding that techniques which previously reported strong improvements for non-pretrained models fail to consistently improve performance for pretrained transformers, even when training data is limited. We hope this empirical analysis helps inform practitioners where data augmentation techniques may confer improvements.

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。

↑