Cross-Domain Generalization Limits of Vision Foundation Models in Facial Deepfake Detection

ArXi:2605.24965v1 Announce Type: cross The rapid evolution of generative models has enabled the creation of hyper-realistic facial deepfakes, exposing a critical vulnerability in modern digital forensics: the inability of detectors to generalize to unseen manipulation techniques. Traditional networks suffer from representation collapse, overfitting to localized artifact fingerprints of specific