Research map / Adversarial robustness and security / Black-box and transfer attacks
TRM-UAP: Enhancing the Transferability of Data-Free Universal Adversarial Perturbation via Truncated Ratio Maximization
Yiran Liu, Xin Feng, Yunlong Wang, Wu Yang, Di Ming
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 |
|---|---|---|---|
| maximize ratio of positive to negative activations instead of positive activation only, with truncation on both (ablation) | TRM-UAP · ImageNet val, ResNet152 and GoogleNet surrogates | fooling rate: around 8~18% FR gain | Improved |
| retain only low-level (shallow) conv layers for activation loss rather than all layers (ablation) | TRM-UAP · ImageNet val, white-box, all five CNNs | fooling rate | Improved |
| use UAP from former 80% layers versus latter 20% layers (ablation) | TRM-UAP · VGG16 label distribution | UAP top-1 label dominance | Improved |
| TRM-UAP versus Cosine-UAP | TRM-UAP · ImageNet val, 5-model average | fooling rate: 84.08 -> 86.39 | Improved |
| TRM-UAP versus Cosine-UAP on GoogleNet | TRM-UAP · ImageNet val, GoogleNet | fooling rate: 87.57 -> 85.32 | Hurt |
| TRM-UAP versus GD-UAP on logit loss | TRM-UAP · ImageNet val, C&W logit loss | logit loss | Improved |
Later papers that compared with this method
1 later accepted paper reports TRM-UAP as a baseline in their results tables; 1 of them report a better number on most of the results they share with it.
- Data-free Universal Adversarial Perturbation with Pseudo-semantic Prior (CVPR 2025 · better on 8 of 9 shared results)
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- Data-free Universal Adversarial Perturbation with Pseudo-semantic Prior (CVPR 2025 · similarity 0.93)
- Improving Generalization of Universal Adversarial Perturbation via Dynamic Maximin Optimization (AAAI 2025 · similarity 0.90)
- Data-Free Universal Attack by Exploiting the Intrinsic Vulnerability of Deep Models (AAAI 2025 · similarity 0.90)
- Generate Universal Adversarial Perturbations for Few-Shot Learning (NeurIPS 2024 · similarity 0.88)
- Vanish into Thin Air: Cross-prompt Universal Adversarial Attacks for SAM2 (NeurIPS 2025 · similarity 0.86)
- Democratic Training Against Universal Adversarial Perturbations (ICLR 2025 · similarity 0.86)
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.
