Adversarial perturbations reliably fabricate membership signals in vision-model MIAs, separated by a gradient-norm collapse trajectory that enables robust detection and inference.
Adversarial weight perturbation helps robust generalization
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
citation-role summary
method 1
citation-polarity summary
fields
cs.CV 2years
2026 2roles
method 1polarities
use method 1representative citing papers
GRACE improves ID accuracy by 10.8% and adversarial accuracy by 13.5% on ImageNet-tuned CLIP while holding OOD accuracy near the zero-shot baseline.
citing papers explorer
-
A Unified Perspective on Adversarial Membership Manipulation in Vision Models
Adversarial perturbations reliably fabricate membership signals in vision-model MIAs, separated by a gradient-norm collapse trajectory that enables robust detection and inference.
-
The Geometry of Robustness: Optimizing Loss Landscape Curvature and Feature Manifold Alignment for Robust Finetuning of Vision-Language Models
GRACE improves ID accuracy by 10.8% and adversarial accuracy by 13.5% on ImageNet-tuned CLIP while holding OOD accuracy near the zero-shot baseline.