1 paper
Dongmei Shan, Wei Lian, Chongxia Wang
Backdoor attacks pose a critical threat to deep learning, especially in safety-sensitive 3D domains such as autonomous driving and robotics. While potent, existing attacks on 3D po…