Modern generative diffusion models rely on vast training datasets, often including images with uncertain ownership or usage rights. Radioactive watermarks -- marks that transfer to a model's outputs -- can help detect when such unauthorized data has been used for training. Moreover, aside from being radioactive, an effective watermark for protecting images from unauthorized training also needs to meet other existing requirements, such as imperceptibility, robustness, and multi-bit capacity. To overcome these challenges, we propose HMARK, a novel multi-bit watermarking scheme, which encodes ownership information as secret bits in the semantic-latent space (h-space) for image diffusion models. By leveraging the interpretability and semantic significance of h-space, ensuring that watermark signals correspond to meaningful semantic attributes, the watermarks embedded by HMARK exhibit radioactivity, robustness to distortions, and minimal impact on perceptual quality. Experimental results demonstrate that HMARK achieves 98.57% watermark detection accuracy, 95.07% bit-level recovery accuracy, 100% recall rate, and 1.0 AUC on images produced by the downstream adversarial model finetuned with LoRA on watermarked data across various types of distortions.
翻译:现代生成式扩散模型依赖于海量训练数据集,这些数据集中常包含所有权或使用权不明确的图像。放射性水印——即能够传递至模型输出的标记——有助于检测此类未经授权数据是否被用于训练。此外,除了具备放射性,用于保护图像免遭未经授权训练的有效水印还需满足其他现有要求,如不可感知性、鲁棒性和多比特容量。为应对这些挑战,我们提出HMARK,一种新颖的多比特水印方案,该方案将所有权信息编码为图像扩散模型语义潜在空间(h空间)中的秘密比特。通过利用h空间的可解释性和语义重要性,确保水印信号对应有意义的语义属性,HMARK嵌入的水印展现出放射性、对失真的鲁棒性以及对感知质量的最小影响。实验结果表明,在经LoRA对水印数据进行微调的下游对抗模型生成的图像上,HMARK在各类失真条件下实现了98.57%的水印检测准确率、95.07%的比特级恢复准确率、100%的召回率以及1.0的AUC值。