 Certifiable Black-Box Attacks with Randomized Adversarial Examples:
Breaking Defenses with Provable ConfidenceConference on Computer and Communications Security (CCS), 2023 |
 Task and Model Agnostic Adversarial Attack on Graph Neural NetworksAAAI Conference on Artificial Intelligence (AAAI), 2021 |
 Dyn-Backdoor: Backdoor Attack on Dynamic Link PredictionIEEE Transactions on Network Science and Engineering (TNSE), 2021 |
 A Targeted Universal Attack on Graph Convolutional NetworkNeural Processing Letters (NPL), 2020 |
 Efficient, Direct, and Restricted Black-Box Graph Evasion Attacks to
Any-Layer Graph Neural Networks via Influence FunctionWeb Search and Data Mining (WSDM), 2020 |
