Adaptive Model Inversion Attacks Generalize a Privacy-Robustness Tradeoff
cs.LG, cs.CR
Submitted: 2026-10-06
Updated: 2026-10-06
Code: https://github.com/LynnHo/HD-CelebA-Cropper
License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Terminology
Sources
- Square Attack: a query-efficient black-box adversarial attack via random search
- Adversarial Robustness as a Prior for Learned Representations
- Deep Residual Learning for Image Recognition
- Model Inversion Robustness: Can Transfer Learning Help?
- Adversarial Examples Are Not Bugs, They Are Features
- A Style-Based Generator Architecture for Generative Adversarial Networks
- Are Perceptually-Aligned Gradients a General Property of Robust Classifiers?
- On the Vulnerability of Skip Connections to Model Inversion Attacks
- Towards Deep Learning Models Resistant to Adversarial Attacks
- Robust or Private? Adversarial Training Makes Models More Vulnerable to Privacy Attacks
- A Closer Look at GAN Priors: Exploiting Intermediate Features for Enhanced Model Inversion Attacks
- Plug & Play Attacks: Towards Robust and Flexible Model Inversion Attacks
- Be Careful What You Smooth For: Label Smoothing Can Be a Privacy Shield but Also a Catalyst for Model Inversion Attacks
- Improving Robustness to Model Inversion Attacks via Mutual Information Regularization
- The Secret Revealer: Generative Model-Inversion Attacks Against Deep Neural Networks
Related papers
- Polynomial-Augmented Neural Networks (PANNs) with Weak Orthogonality Constraints for Enhanced Function and PDE Approximation
- AIRL-S: Unifying Reinforcement Learning and Search-Based Test-Time Scaling via Adversarial Inverse Reinforcement Learning
- Transformers as Bayesian In-Context Experimenters: Smoothness-Adaptive Efficient ATE Estimation
- Convergence issues in Relational Concept Analysis based on AOC-posets
- Beliefs Beyond Posteriors: The Predictive Cost of ELBO Binding
- Understanding Diffusion Models via Ratio-Based Function Approximation with SignReLU Networks