Goto

Collaborating Authors

 Asia





Breaking the False Sense of Security in Backdoor Defense through Re-Activation Attack

Neural Information Processing Systems

To further verify this finding, we empirically show that these dormant backdoors can be easily re-activated during inference stage, by manipulating the original trigger with well-designed tiny perturbation using universal adversarial attack.



Agent Planning with World Knowledge Model

Neural Information Processing Systems

Imitating humans' mental world knowledge model which provides global prior knowledge before the task and maintains local dynamic knowledge during the task, in this paper, we introduce parametric W orld K nowledge M odel ( WKM) to facilitate agent