Research map / Adversarial robustness and security / Attacks on text and graphs
Pseudo Label-Guided Model Inversion Attack via Conditional Generative Adversarial Network
Xiaojian Yuan, Kejiang Chen, Jie Zhang, Weiming Zhang, Nenghai Yu, Yang Zhang
Read the paperCheck your idea against 137,217 papers
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 |
|---|---|---|---|
| Use PLG-MI instead of KED-MI | Face recognition inversion · FaceScrub → CelebA, ResNet-152 | Attack Acc: .15±.0011 -> .68±.0020 | Improved |
| Use PLG-MI instead of KED-MI | Face recognition inversion · FFHQ → CelebA, ResNet-152 | Attack Acc: .74±.0028 -> 1.±.0000 | Improved |
| Use PLG-MI instead of KED-MI | Model inversion · MNIST | Attack Acc: .29±.0228 -> .60±.0465 | Improved |
| Use PLG-MI instead of KED-MI | Model inversion · CIFAR-10 | Attack Acc: .50±.0392 -> .76±.0318 | Improved |
| Replace cross-entropy with max-margin loss (ablation) | PLG-MI reconstruction · FaceScrub → CelebA | Attack Acc | Improved |
| Vary top-n selection size (ablation) | PLG-MI pseudo-label generation · FaceScrub or FFHQ → CelebA | Attack performance | No clear effect |
| Set the generator constraint strength to α=0.2 (ablation) | PLG-MI cGAN · FaceScrub → CelebA | Attack Acc and FID | Improved |
| Use random augmentations during reconstruction (ablation) | PLG-MI latent search · FaceScrub → CelebA | Attack Acc and visual quality | Improved |
| Use more than two reconstruction augmentations (ablation) | PLG-MI latent search · FaceScrub → CelebA | Attack performance | No clear effect |
Closest papers by meaning
- A Closer Look at GAN Priors: Exploiting Intermediate Features for Enhanced Model Inversion Attacks (ECCV 2024 · similarity 0.92)
- The Secret Revealer: Generative Model-Inversion Attacks Against Deep Neural Networks (CVPR 2020 · similarity 0.91)
- Pseudo-Private Data Guided Model Inversion Attacks (NeurIPS 2024 · similarity 0.89)
- A Sample-Level Evaluation and Generative Framework for Model Inversion Attacks (AAAI 2025 · similarity 0.89)
- Attack Deterministic Conditional Image Generative Models for Diverse and Controllable Generation (AAAI 2024 · similarity 0.86)
- Fusing Conditional Submodular GAN and Programmatic Weak Supervision (AAAI 2024 · 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.
