The scarcity of labeled dataset leads us to consider the integration of data augmentation and active learning.

Whileprevious work tackles this issue by using explicit labeling on the spuriously correlated attributes or presuming a particular bias type, we instead utilize a cheaper, yet generic form of human knowledge, which can be widely applicable to various types ofbias.

Still, many relevant tasks such as object manipulation pose significant challenges forRL.

Such a weak generalization capability hinders the applicability of current deepfake detectors.