面向大语言模型的主动隐私遗忘：在几乎不影响模型效用的情况下保护个人身份信息
Proactive Privacy Amnesia for Large Language Models: Safeguarding PII with Negligible Impact on Model Utility
AI总结：
本文提出主动隐私遗忘（PPA）方法，通过主动识别并遗忘与PII最相关的关键记忆，再植入替代记忆，在几乎不影响模型效用的情况下有效保护大语言模型中的个人身份信息。
摘要：
随着大语言模型（LLMs）的兴起，越来越多的研究已经认识到它们在恶意攻击下泄露个人身份信息（PII）的风险。尽管已有工作致力于保护LLMs中的PII，但现有方法难以在隐私保护与维持模型效用之间取得平衡。本文受认知科学中遗忘研究的启发，提出了一种新颖的方法——主动隐私遗忘（Proactive Privacy Amnesia，PPA），用于在保护LLMs中PII的同时保持其效用。该机制的工作原理是主动识别并遗忘序列中与PII关联最密切的关键记忆，随后通过植入合适的替代记忆来维持LLM的功能。我们在多个模型上进行了评估，以保护常见PII（如电话号码和物理地址）免受当前流行的针对PII的攻击，证明了我们的方法相比其他现有防御技术的优越性。结果表明，我们的PPA方法完全消除了电话号码泄露风险，降幅达100%，并将物理地址泄露风险显著降低了9.8%至87.6%，同时保持了可比的模型效用性能。
英文摘要：
With the rise of large language models (LLMs), increasing research has recognized their risk of leaking personally identifiable information (PII) under malicious attacks. Although efforts have been made to protect PII in LLMs, existing methods struggle to balance privacy protection with maintaining model utility. In this paper, inspired by studies of amnesia in cognitive science, we propose a novel approach, Proactive Privacy Amnesia (PPA), to safeguard PII in LLMs while preserving their utility. This mechanism works by actively identifying and forgetting key memories most closely associated with PII in sequences, followed by a memory implanting using suitable substitute memories to maintain the LLM's functionality. We conduct evaluations across multiple models to protect common PII, such as phone numbers and physical addresses, against prevalent PII-targeted attacks, demonstrating the superiority of our method compared with other existing defensive techniques. The results show that our PPA method completely eliminates the risk of phone number exposure by 100% and significantly reduces the risk of physical address exposure by 9.8% - 87.6%, all while maintaining comparable model utility performance.
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。
