Generative models have enabled the creation of contents that are indistinguishable from those taken from the nature. Open-source development of such models raised concerns about the risks in their misuse for malicious purposes. One potential risk mitigation strategy is to attribute generative models via fingerprinting. Current fingerprinting methods exhibit significant tradeoff between robust attribution accuracy and generation quality, and also lack designing principles to improve this tradeoff. This paper investigates the use of latent semantic dimensions as fingerprints, from where we can analyze the effects of design variables, including the choice of fingerprinting dimensions, strength, and capacity, on the accuracy-quality tradeoff. Compared with previous SOTA, our method requires minimum computation and is more applicable to large-scale models. We use StyleGAN2 and the latent diffusion model to demonstrate the efficacy of our method.
翻译:生成模型使得我们可以创造出与自然物品无法区分的内容。这些模型的开源开发引起了人们对于其被滥用的担忧。为了缓解这种风险,一种潜在的策略就是通过指纹的方式来对生成模型进行追踪。目前的指纹法在准确性和生成质量之间存在着巨大的平衡点,同时也缺乏设计准则以优化这种平衡。本文旨在探究利用潜在语义维度作为指纹的可能性,并且研究了指纹的设计变量,包括指纹维度、强度和容量,三者对于准确度和质量的平衡点的影响。相比于以前的发展水平,我们的方法计算量较小,更适用于大规模的生成模型。我们以StyleGAN2和潜在扩散模型为例,证明了我们方法的有效性。