all InfoSec news
Are Diffusion Models Vulnerable to Membership Inference Attacks?. (arXiv:2302.01316v1 [cs.CV])
cs.CR updates on arXiv.org arxiv.org
Diffusion-based generative models have shown great potential for image
synthesis, but there is a lack of research on the security and privacy risks
they may pose. In this paper, we investigate the vulnerability of diffusion
models to Membership Inference Attacks (MIAs), a common privacy concern. Our
results indicate that existing MIAs designed for GANs or VAE are largely
ineffective on diffusion models, either due to inapplicable scenarios (e.g.,
requiring the discriminator of GANs) or inappropriate assumptions (e.g., closer
distances between …
attacks diffusion models gans generative great may privacy privacy concern privacy risks research results risks security vulnerability vulnerable