arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2309.09697cs.CL
收藏

通过考虑所有标签评估预训练语言模型在自然语言推理中的性别偏见

Evaluating Gender Bias of Pre-trained Language Models in Natural Language Inference by Considering All Labels

Tokyo Institute of Technology(东京工业大学)

Panatchakorn Anantaprayoon, Masahiro Kaneko, Naoaki Okazaki

AI总结:

针对现有NLI偏见评估仅关注单类标签的缺陷,提出考虑全部三类标签的NLI-CoAL方法,经元评估验证准确性优于基线,且适配英、日、中多语言,还首次构建了日中两种语言的NLI偏见评估数据集。

摘要:

针对多种语言的预训练语言模型(Pre-trained Language Models, PLMs)中已被发现存在歧视性性别偏见。在自然语言推理(Natural Language Inference, NLI)任务中,现有的偏见评估方法仅关注三类标签中某一特定标签(如中性标签)的预测结果,但这类评估方法可能存在偏差,因为不同的偏见推理对应着不同的预测标签。针对这一局限,我们提出了一种面向PLMs的偏见评估方法NLI-CoAL,该方法会考量NLI任务的全部三类标签。首先,我们构建了三组分别对应不同偏见类型的评估数据组;随后,基于每组数据对应的标签输出定义了偏见度量指标。实验中,我们引入了一种用于NLI偏见度量的元评估技术,经该技术验证,我们的偏见度量指标相比基线方法能更好地区分带偏见的错误推理与无偏见的错误推理,可实现更准确的偏见评估。我们构建了英语、日语和中文三种语言的数据集,成功验证了该偏见度量指标在多语言间的兼容性。最后,我们观测了不同语言PLMs中的偏见倾向。据我们所知,我们是首个构建日语和中文NLI评估数据集、并从NLI维度度量PLMs偏见的研究。

英文摘要:

Discriminatory gender biases have been found in Pre-trained Language Models (PLMs) for multiple languages. In Natural Language Inference (NLI), existing bias evaluation methods have focused on the prediction results of one specific label out of three labels, such as neutral. However, such evaluation methods can be inaccurate since unique biased inferences are associated with unique prediction labels. Addressing this limitation, we propose a bias evaluation method for PLMs, called NLI-CoAL, which considers all the three labels of NLI task. First, we create three evaluation data groups that represent different types of biases. Then, we define a bias measure based on the corresponding label output of each data group. In the experiments, we introduce a meta-evaluation technique for NLI bias measures and use it to confirm that our bias measure can distinguish biased, incorrect inferences from non-biased incorrect inferences better than the baseline, resulting in a more accurate bias evaluation. We create the datasets in English, Japanese, and Chinese, and successfully validate the compatibility of our bias measure across multiple languages. Lastly, we observe the bias tendencies in PLMs of different languages. To our knowledge, we are the first to construct evaluation datasets and measure PLMs' bias from NLI in Japanese and Chinese.

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。

↑