We are excited to announce that the paper "Attacking Graph Neural Networks with Bit Flips: Weisfeiler and Leman Go Indifferent" by Lorenz Kummer, Samir Moustafa, Sebastian Schrittwieser, Nils Kriege, and Wilfried Gansterer has been accepted at the SIGKDD Conference on Knowledge Discovery and Data Mining (KDD 2024) with an acceptance rate of ~20%.
Have a look at our paper to learn how bit-flip attacks for graph neural networks can be designed that degrade their ability to distinguish graph structure.