arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2406.04892cs.CL
收藏

在数据精简下的性别歧视检测

Sexism Detection on a Data Diet

GESIS Leibniz Institute for the Social Sciences(GESIS莱布尼茨社会科学研究所);Centro Singular de Investigación en Tecnoloxías Intelixentes (CiTIUS), Universidade de Santiago de Compostela(智能技术 singular 研究中心(CiTIUS),圣地亚哥德孔波斯特拉大学)

Rabiraj Bandyopadhyay, Dennis Assenmacher, Jose M. Alonso Moral, Claudia Wagner

AI总结:

本文研究利用影响分数进行数据剪枝以训练性别歧视检测模型,发现大量数据可被移除而不显著降低性能,但现有剪枝策略会加剧类别不平衡,甚至导致仇恨类别完全缺失。

摘要:

随着社交媒体使用的增加,网络仇恨内容的扩散也在加剧。相应地,基于自然语言处理和深度学习方法构建的、旨在识别有害文本内容的自动化工具也取得了显著进展。尽管已知训练深度学习模型需要大量标注数据,但最近的一系列研究表明,在特定数据子集上训练的模型仍能保持与在完整数据集上训练的模型相当的性能。在这项工作中,我们展示了如何在训练模型时利用影响分数来估计数据点的重要性,并设计了一种应用于性别歧视检测场景的剪枝策略。我们在三个域外数据集上评估了使用不同剪枝策略修剪后的数据所训练模型的性能,并发现与其他工作一致,可以移除大量实例而不会造成显著的性能下降。然而,我们还发现,先前在自然语言推理任务中取得成功的剪枝策略并不能直接适用于有害内容的检测,反而会进一步加剧本就普遍存在的类别不平衡,在最坏的情况下导致仇恨类别的完全缺失。

英文摘要:

There is an increase in the proliferation of online hate commensurate with the rise in the usage of social media. In response, there is also a significant advancement in the creation of automated tools aimed at identifying harmful text content using approaches grounded in Natural Language Processing and Deep Learning. Although it is known that training Deep Learning models require a substantial amount of annotated data, recent line of work suggests that models trained on specific subsets of the data still retain performance comparable to the model that was trained on the full dataset. In this work, we show how we can leverage influence scores to estimate the importance of a data point while training a model and designing a pruning strategy applied to the case of sexism detection. We evaluate the model performance trained on data pruned with different pruning strategies on three out-of-domain datasets and find, that in accordance with other work a large fraction of instances can be removed without significant performance drop. However, we also discover that the strategies for pruning data, previously successful in Natural Language Inference tasks, do not readily apply to the detection of harmful content and instead amplify the already prevalent class imbalance even more, leading in the worst-case to a complete absence of the hateful class.

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。

↑