Fetching the paper…

Beyond Gradients: Exploiting Adversarial Priors in Model Inversion Attacks · Around