ForgetMe：评估生成模型中的选择性遗忘
ForgetMe: Evaluating Selective Forgetting in Generative Models
AI总结：
针对扩散模型选择性遗忘难题，提出自动数据集创建框架与Entangled评估指标，构建ForgetMe基准，结合LoRA微调实现隐私保护。
摘要：
扩散模型在图像生成中的广泛采用增加了对符合隐私要求的机器遗忘的需求。然而，由于扩散模型的高维特性和复杂的特征表示，实现选择性遗忘仍然具有挑战性，因为现有方法在移除敏感信息的同时难以保持非敏感区域的一致性。为了解决这一问题，我们提出了一种基于提示分层编辑和无训练局部特征移除的自动数据集创建框架，构建了ForgetMe数据集，并引入了Entangled评估指标。Entangled指标通过评估目标区域与背景区域之间的相似性和一致性来量化遗忘效果，并支持成对（Entangled-D）和非成对（Entangled-S）图像数据，从而实现无监督评估。ForgetMe数据集涵盖了多样化的真实和合成场景，包括CUB-200-2011（鸟类）、Stanford-Dogs、ImageNet以及一个合成猫数据集。我们在Stable Diffusion上应用LoRA微调以实现对该数据集的选择性遗忘，并验证了ForgetMe数据集和Entangled指标的有效性，将它们确立为选择性遗忘的基准。我们的工作为推进隐私保护的生成式人工智能提供了一种可扩展且适应性强的解决方案。
英文摘要：
The widespread adoption of diffusion models in image generation has increased the demand for privacy-compliant unlearning. However, due to the high-dimensional nature and complex feature representations of diffusion models, achieving selective unlearning remains challenging, as existing methods struggle to remove sensitive information while preserving the consistency of non-sensitive regions. To address this, we propose an Automatic Dataset Creation Framework based on prompt-based layered editing and training-free local feature removal, constructing the ForgetMe dataset and introducing the Entangled evaluation metric. The Entangled metric quantifies unlearning effectiveness by assessing the similarity and consistency between the target and background regions and supports both paired (Entangled-D) and unpaired (Entangled-S) image data, enabling unsupervised evaluation. The ForgetMe dataset encompasses a diverse set of real and synthetic scenarios, including CUB-200-2011 (Birds), Stanford-Dogs, ImageNet, and a synthetic cat dataset. We apply LoRA fine-tuning on Stable Diffusion to achieve selective unlearning on this dataset and validate the effectiveness of both the ForgetMe dataset and the Entangled metric, establishing them as benchmarks for selective unlearning. Our work provides a scalable and adaptable solution for advancing privacy-preserving generative AI.
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。
