Calibrating Uncertainty for Zero-Shot Adversarial CLIP

ArXi:2512.12997v2 Announce Type: replace-cross CLIP delivers strong zero-shot classification but remains highly vulnerable to adversarial attacks. Prior adversarial fine-tuning work primarily matches predicted logits between clean and adversarial examples, which overlooks uncertainty calibration and may degrade the zero-shot generalization. A common expectation in reliable uncertainty estimation is that predictive uncertainty should increase as inputs become difficult or shift away from the.