Casey Gollan ♥ liked @Riedl@sigmoid.social 3 years ago M Mark Riedl @Riedl@sigmoid.social · 3y 🐘 Stable Diffusion can memorize and output impages from its training set, including copyrighted content https://arxiv.org/abs/2301.13188arXiv.orgExtracting Training Data from Diffusion ModelsImage diffusion models such as DALL-E 2, Imagen, and Stable Diffusion have attracted significant attention due to their ability to generate high-quality synthetic images. In this work, we show that diffusion models memorize individual images from their training data and emit them at generation time. With a generate-and-filter pipeline, we extract over a thousand training examples from state-of-the-art models, ranging from photographs of individual people to trademarked company logos. We also train hundreds of diffusion models in various settings to analyze how different modeling and data decisions affect privacy. Overall, our results show that diffusion models are much less private than prior generative models such as GANs, and that mitigating these vulnerabilities may require new advances in privacy-preserving training. Reply Like Repost Loading webmentions…
M Mark Riedl @Riedl@sigmoid.social · 3y 🐘 Stable Diffusion can memorize and output impages from its training set, including copyrighted content https://arxiv.org/abs/2301.13188arXiv.orgExtracting Training Data from Diffusion ModelsImage diffusion models such as DALL-E 2, Imagen, and Stable Diffusion have attracted significant attention due to their ability to generate high-quality synthetic images. In this work, we show that diffusion models memorize individual images from their training data and emit them at generation time. With a generate-and-filter pipeline, we extract over a thousand training examples from state-of-the-art models, ranging from photographs of individual people to trademarked company logos. We also train hundreds of diffusion models in various settings to analyze how different modeling and data decisions affect privacy. Overall, our results show that diffusion models are much less private than prior generative models such as GANs, and that mitigating these vulnerabilities may require new advances in privacy-preserving training.