Research map / Adversarial robustness and security / Certified and empirical robustness
Dual-branch Robust Unlearnable Examples
Xianlong Wang, Hangtao Zhang, Wenbo Pan, Ziqi Zhou, Changsong Jiang, Li Zeng, Xiaohua Jia
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 |
|---|---|---|---|
| add unlearnability-enhancing ensemble (UEE) to spatial+color branches (ablation) | DUNE · CIFAR-10 ResNet18 | average test accuracy over defenses (%): 31.15 -> 19.67 | Improved |
| add UEE to spatial+color branches (ablation) | DUNE · CIFAR-10 VGG19 | average test accuracy over defenses (%): 27.11 -> 18.09 | Improved |
| add color branch to spatial branch plus ensemble (ablation) | DUNE · CIFAR-10 ResNet18 | average test accuracy over defenses (%): 50.20 -> 19.67 | Improved |
| add spatial branch to color branch plus ensemble (ablation) | DUNE · CIFAR-10 ResNet18 | average test accuracy over defenses (%): 32.68 -> 19.67 | Improved |
| add color and spatial branches and ensemble to spatial-only (ablation) | DUNE · CIFAR-10 ResNet18 | average test accuracy over defenses (%): 57.05 -> 19.67 | Improved |
| add spatial branch and ensemble to color-only (ablation) | DUNE · CIFAR-10 ResNet18 | average test accuracy over defenses (%): 43.67 -> 19.67 | Improved |
| add color branch to spatial branch under ISS-J (ablation) | DUNE (vs spatial + Region-4) · CIFAR-10 ResNet18 | test accuracy under ISS-J (%): 80.97 -> 28.88 | Improved |
| add DUNE instead of best baseline CUDA under no-defense/ISS-J average | DUNE · CIFAR-10 ResNet18 | average test accuracy (%): 39.58 -> 23.29 | Improved |
| use PSO color-branch optimization instead of PGD | DUNE · CIFAR-10 | total generation time (hrs): 0.3459 -> 0.9574 | Hurt |
| use PSO color-branch optimization instead of PGD | DUNE · ImageNet-100 | total generation time (hrs): 1.3641 -> 7.5766 | Hurt |
| increase beta excessively (ablation) | DUNE spatial branch · CIFAR-10 ResNet18 | test accuracy | Hurt |
| apply DUNE under ISS-J instead of EM | DUNE · CIFAR-10 ResNet18 | test accuracy under ISS-J (%): 78.91 -> 28.88 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- CUDA — CUDA: Convolution-Based Unlearnable Datasets (CVPR 2023) · better on 3 of 4 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
- Stable Unlearnable Example: Enhancing the Robustness of Unlearnable Examples via Stable Error-Minimizing Noise (AAAI 2024 · similarity 0.87)
- How Far Are We from True Unlearnability? (ICLR 2025 · similarity 0.86)
- Purify Unlearnable Examples via Rate-Constrained Variational Autoencoders (ICML 2024 · similarity 0.85)
- Toward Improving Robustness and Accuracy in Unsupervised Domain Adaptation (AAAI 2025 · similarity 0.85)
- Detecting and Corrupting Convolution-based Unlearnable Examples (AAAI 2025 · similarity 0.84)
- Improving robustness to corruptions with multiplicative weight perturbations (NeurIPS 2024 · similarity 0.84)
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.
