NLPGuard：一个用于缓解NLP分类器使用受保护属性的框架
NLPGuard: A Framework for Mitigating the Use of Protected Attributes by NLP Classifiers
AI总结：
针对NLP分类器依赖受保护属性的问题，提出NLPGuard框架，通过修改训练数据集在不损失准确性的前提下将依赖性降低高达79%。
摘要：
AI法规预计将禁止机器学习模型在训练期间使用敏感属性。然而，依赖深度学习的最新自然语言处理（NLP）分类器作为黑盒系统运行，使此类滥用的检测和补救变得复杂。NLP中的传统偏见缓解方法旨在基于性别或种族等属性实现不同群体间的可比性能，但未能解决依赖受保护属性的根本问题。为了部分解决这一问题，我们引入了NLPGuard，这是一个用于缓解NLP分类器对受保护属性依赖的框架。NLPGuard将未标记数据集、现有的NLP分类器及其训练数据作为输入，生成一个修改后的训练数据集，该数据集在不损害准确性的情况下显著减少了对受保护属性的依赖。NLPGuard被应用于三个分类任务：识别有毒语言、情感分析和职业分类。我们的评估表明，当前的NLP分类器严重依赖受保护属性，高达$23\%$的最具预测性的单词与这些属性相关。然而，NLPGuard有效地将这种依赖性降低了高达$79\%$，同时略微提高了准确性。
英文摘要：
AI regulations are expected to prohibit machine learning models from using sensitive attributes during training. However, the latest Natural Language Processing (NLP) classifiers, which rely on deep learning, operate as black-box systems, complicating the detection and remediation of such misuse. Traditional bias mitigation methods in NLP aim for comparable performance across different groups based on attributes like gender or race but fail to address the underlying issue of reliance on protected attributes. To partly fix that, we introduce NLPGuard, a framework for mitigating the reliance on protected attributes in NLP classifiers. NLPGuard takes an unlabeled dataset, an existing NLP classifier, and its training data as input, producing a modified training dataset that significantly reduces dependence on protected attributes without compromising accuracy. NLPGuard is applied to three classification tasks: identifying toxic language, sentiment analysis, and occupation classification. Our evaluation shows that current NLP classifiers heavily depend on protected attributes, with up to $23\%$ of the most predictive words associated with these attributes. However, NLPGuard effectively reduces this reliance by up to $79\%$, while slightly improving accuracy.
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。
