arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2302.09345cs.CL
收藏

提高语言模型的分布外泛化能力:反事实扩充数据是不够的

Improving the Out-Of-Distribution Generalization Capability of Language Models: Counterfactually-Augmented Data is not Enough

Shanghai Jiao Tong University(上海交通大学);Sun Yat-sen University(中山大学)

Caoyun Fan, Wenqing Chen, Jidong Tian, Yitian Li, Hao He, Yaohui Jin

摘要:

反事实增强数据(CAD)可以提高语言模型的分布外(OOD)泛化能力,因为CAD可以引导语言模型利用因果特征并排除虚假相关。然而,面向对象的CAD综合的经验结果并不像预期的那样有效。在本文中,我们归因于CAD引起的近视现象:语言模型只关注在增强中被编辑的因果特征,而排除其他未编辑的因果特征。因此,CAD的潜力没有得到充分利用。基于CAD的结构特性,设计了两个附加约束,帮助语言模型提取CAD中包含的更完整的因果特征,从而提高了面向对象设计的泛化能力。我们在情感分析和自然语言推理两个任务上对我们的方法进行了评估,实验结果表明,我们的方法可以释放CAD的潜力,提高语言模型的OOD泛化能力。

英文摘要:

Counterfactually-Augmented Data (CAD) has the potential to improve language models' Out-Of-Distribution (OOD) generalization capability, as CAD induces language models to exploit causal features and exclude spurious correlations. However, the empirical results of OOD generalization on CAD are not as efficient as expected. In this paper, we attribute the inefficiency to Myopia Phenomenon caused by CAD: language models only focus on causal features that are edited in the augmentation and exclude other non-edited causal features. As a result, the potential of CAD is not fully exploited. Based on the structural properties of CAD, we design two additional constraints to help language models extract more complete causal features contained in CAD, thus improving the OOD generalization capability. We evaluate our method on two tasks: Sentiment Analysis and Natural Language Inference, and the experimental results demonstrate that our method could unlock CAD's potential and improve language models' OOD generalization capability.

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。

↑