In this work, we address the problem of text anonymization where the goal is to prevent adversaries from correctly inferring private attributes of the author, while keeping the text utility, i.e., meaning and semantics. We propose IncogniText, a technique that anonymizes the text to mislead a potential adversary into predicting a wrong private attribute value. Our empirical evaluation shows a reduction of private attribute leakage by more than 90% across 8 different private attributes. Finally, we demonstrate the maturity of IncogniText for real-world applications by distilling its anonymization capability into a set of LoRA parameters associated with an on-device model. Our results show the possibility of reducing privacy leakage by more than half with limited impact on utility.

翻译：本文致力于解决文本匿名化问题，其目标是在保持文本效用（即意义与语义）的同时，防止攻击者正确推断作者的私有属性。我们提出了IncogniText技术，该技术通过匿名化文本误导潜在攻击者预测错误的私有属性值。实证评估表明，该方法在8种不同私有属性上的泄露率降低了90%以上。最后，我们通过将IncogniText的匿名化能力提炼为一组与端侧模型关联的LoRA参数，证明了其在实际应用中的成熟性。结果表明，该方法能够在有限影响文本效用的前提下，将隐私泄露风险降低一半以上。