评估基于RoBERTa的仇恨言论检测模型中的简单去偏技术
Evaluating Simple Debiasing Techniques in RoBERTa-based Hate Speech Detection Models
AI总结：
本文针对基于RoBERTa的仇恨言论检测模型中对非裔美国人英语方言的误分类偏见，应用并评估了简单去偏技术，发现其效果高度依赖训练数据集构建方法，妥善考虑表征偏差可减少方言子群间的检测差异。
摘要：
仇恨言论检测任务因训练所用底层数据集中存在标注偏差，已知会对非裔美国人英语（AAE）方言文本产生偏见。这导致了一种差异：与非AAE文本相比，正常的AAE文本更可能被误分类为辱骂/仇恨言论。过去已开发出简单的去偏技术来应对此类差异，在本研究中，我们在基于RoBERTa的编码器范围内应用并评估这些技术。实验结果表明，这些技术的成功在很大程度上取决于训练数据集构建所使用的方法，但若妥善考虑表征偏差，它们能减少仇恨言论检测任务中方言子群间的差异。
英文摘要：
The hate speech detection task is known to suffer from bias against African American English (AAE) dialect text, due to the annotation bias present in the underlying hate speech datasets used to train these models. This leads to a disparity where normal AAE text is more likely to be misclassified as abusive/hateful compared to non-AAE text. Simple debiasing techniques have been developed in the past to counter this sort of disparity, and in this work, we apply and evaluate these techniques in the scope of RoBERTa-based encoders. Experimental results suggest that the success of these techniques depends heavily on the methods used for training dataset construction, but with proper consideration of representation bias, they can reduce the disparity seen among dialect subgroups on the hate speech detection task.
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。
