-
Membership-Invariant Subspace Training
Membership-Invariant Subspace Training (MIST) is a method for training classifiers that acts as a defense designed to specifically defend against black-box membership inference... -
CIFAR-10, CIFAR-100, STL-10, and Tiny-ImageNet
The dataset used in the paper is a de-noising diffusion probabilistic model (DDPM) trained on CIFAR-10, CIFAR-100, STL-10, and Tiny-ImageNet. -
Various Datasets
The datasets used in the paper are described as follows: WikiMIA, BookMIA, Temporal Wiki, Temporal arXiv, ArXiv-1 month, Multi-Webdata, LAION-MI, Gutenberg.