Research map / Adversarial robustness and security / Attacks on text and graphs
Learning to Generate Inversion-Resistant Model Explanations
Hoyong Jeong, Suyoung Lee, Sung Ju Hwang, Sooel Son
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| apply GNIME to CelebA-LRP explanations | ExpMI attack · CelebA-LRP | SSIM: .8233 -> .4960 | Improved |
| apply GNIME to CelebA-Grad explanations | ExpMI attack · CelebA-Grad | DeePSiM: .2470 -> .1880 | Improved |
| apply Gaussian noise (RND) to CelebA-LRP explanations | ExpMI attack · CelebA-LRP | SSIM: .8233 -> .6370 | Improved |
| adversary trains on FaceScrub instead of CelebA | GNIME defense · CelebA-LRP | TCA: 0.2086 -> 0.0012 | Improved |
| apply GNIME with gamma 0.0040 (ablation) | GNIME · CelebA explanation functionality | DeePSiM: 0.6340 | Mixed |
| apply GNIME to attribute prediction from reconstructions | ExpMI attack · CelebA attribute classifiers | F1: 0.7672 -> 0.3048 | Improved |
| use reverse sigmoid perturbation against model-extraction ExpMI-3 | ExpMI-3 attack · CelebA | SSIM: 0.4928 | Improved |
| apply GNIME noise to explanations | explanation functionality · CelebA-LRP | DeePSiM: 0.6734 | Mixed |
Closest papers by meaning
- The Secret Revealer: Generative Model-Inversion Attacks Against Deep Neural Networks (CVPR 2020 · similarity 0.91)
- Trap-MID: Trapdoor-based Defense against Model Inversion Attacks (NeurIPS 2024 · similarity 0.86)
- A Sample-Level Evaluation and Generative Framework for Model Inversion Attacks (AAAI 2025 · similarity 0.86)
- REFINE: Inversion-Free Backdoor Defense via Model Reprogramming (ICLR 2025 · similarity 0.85)
- Efficient Model Stealing Defense with Noise Transition Matrix (CVPR 2024 · similarity 0.85)
- Theoretical Insights in Model Inversion Robustness and Conditional Entropy Maximization for Collaborative Inference Systems (CVPR 2025 · similarity 0.85)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
