使用少样本遗忘从文本到图像扩散模型中擦除概念
Erasing Concepts from Text-to-Image Diffusion Models with Few-shot Unlearning
AI总结：
提出一种基于少样本遗忘的文本编码器更新方法，通过隐式转换为模型固有潜在概念，在10秒内快速擦除文本到图像扩散模型中的特定概念，速度提升数十至数百倍且效果更自然。
摘要：
由于扩散模型的规模扩大以及视觉和语言领域的进步，从文本生成图像变得更加容易。这些模型使用来自互联网的海量数据进行训练。因此，它们通常包含不受欢迎的内容，例如受版权保护的材料。由于移除此类数据并重新训练模型具有挑战性，因此研究者们已经探索了从预训练模型中擦除特定概念的方法。我们提出了一种新颖的概念擦除方法，该方法使用少样本遗忘更新文本编码器，其中使用了少量真实图像。关于擦除概念后生成图像的讨论一直比较缺乏。虽然存在指定概念转换目标的方法，但指定概念的有效性尚不明确。我们的方法通过转换为模型或图像中固有的潜在概念，隐式地实现了这一点。我们的方法可以在10秒内擦除一个概念，使概念擦除比以往任何时候都更容易实现。隐式转换为相关概念会导致更自然的概念擦除。我们将所提出的方法应用于各种概念，并证实概念擦除可以比现有方法快数十到数百倍。通过改变要更新的参数，我们获得了结果表明，与之前的研究一样，知识主要积累在文本编码器的前馈网络中。我们的代码可在 https://github.com/fmp453/few-shot-erasing 获取
英文摘要：
Generating images from text has become easier because of the scaling of diffusion models and advancements in the field of vision and language. These models are trained using vast amounts of data from the Internet. Hence, they often contain undesirable content such as copyrighted material. As it is challenging to remove such data and retrain the models, methods for erasing specific concepts from pre-trained models have been investigated. We propose a novel concept-erasure method that updates the text encoder using few-shot unlearning in which a few real images are used. The discussion regarding the generated images after erasing a concept has been lacking. While there are methods for specifying the transition destination for concepts, the validity of the specified concepts is unclear. Our method implicitly achieves this by transitioning to the latent concepts inherent in the model or the images. Our method can erase a concept within 10 s, making concept erasure more accessible than ever before. Implicitly transitioning to related concepts leads to more natural concept erasure. We applied the proposed method to various concepts and confirmed that concept erasure can be achieved tens to hundreds of times faster than with current methods. By varying the parameters to be updated, we obtained results suggesting that, like previous research, knowledge is primarily accumulated in the feed-forward networks of the text encoder. Our code is available at \url{https://github.com/fmp453/few-shot-erasing}
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。
