用稀疏自编码器实现扩散模型中概念的可解释删除
SAeUron: Interpretable Concept Unlearning in Diffusion Models with Sparse Autoencoders
- 基于扩散模型各去噪步骤的激活特征，训练稀疏自编码器提取可解释概念
- 在UnlearnCanvas上优于现有方法，能有效消除裸露内容且支持多概念同时移除
- 方法透明可追溯，对抗攻击下仍能防止生成有害内容，适合安全敏感场景
扩散模型虽强大，但可能生成有害或不想要的内容，引发伦理与安全问题。现有机器遗忘方法常缺乏透明性，难以理解对基模型的影响。本文提出SAeUron，利用稀疏自编码器（SAEs）学习扩散模型多去噪步骤的激活特征，捕捉对应具体概念的稀疏可解释特征。在此基础上，设计特征选择机制，精准干预模型激活以阻断目标内容，同时保持整体性能。评估显示，SAeUron在UnlearnCanvas基准上于概念与风格遗忘任务表现更优，且在I2P测试中有效消除裸露内容。单个SAE即可同时移除多个概念，且相比其他方法，显著降低对抗攻击下生成不良内容的风险。代码与检查点已开源。
原文摘要 · Abstract (English)
Diffusion models, while powerful, can inadvertently generate harmful or undesirable content, raising significant ethical and safety concerns. Recent machine unlearning approaches offer potential solutions but often lack transparency, making it difficult to understand the changes they introduce to the base model. In this work, we introduce SAeUron, a novel method leveraging features learned by sparse autoencoders (SAEs) to remove unwanted concepts in text-to-image diffusion models. First, we demonstrate that SAEs, trained in an unsupervised manner on activations from multiple denoising timesteps of the diffusion model, capture sparse and interpretable features corresponding to specific concepts. Building on this, we propose a feature selection method that enables precise interventions on model activations to block targeted content while preserving overall performance. Our evaluation shows that SAeUron outperforms existing approaches on the UnlearnCanvas benchmark for concepts and style unlearning, and effectively eliminates nudity when evaluated with I2P. Moreover, we show that with a single SAE, we can remove multiple concepts simultaneously and that in contrast to other methods, SAeUron mitigates the possibility of generating unwanted content under adversarial attack. Code and checkpoints are available at https://github.com/cywinski/SAeUron.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。