Securing GNNs: Explanation-Based Identification of Backdoored Training Graphs

Open in new window