Research map / Adversarial robustness and security / Attacks on text and graphs
Zero-Sacrifice Persistent-Robustness Adversarial Defense for Pre-Trained Encoders
Zhuxin Lei, Ziyuan Yang, Yi 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 |
|---|---|---|---|
| add adversarial fine-tuning to the dual-branch encoder (ablation) | ZePAD · ANIMALS10 (CIFAR10-pretrained DINO) | robust accuracy: 59.40 -> 74.71 | Improved |
| add the MPAE branch (ablation) | ZePAD · ANIMALS10 (CIFAR10-pretrained DINO) | robust accuracy: 70.35 -> 74.71 | Improved |
| add the BMP branch (ablation) | ZePAD · ANIMALS10 (CIFAR10-pretrained DINO) | benign accuracy: 84.51 -> 91.66 | Improved |
| add the BMP branch (ablation) | ZePAD · ANIMALS10 (CIFAR10-pretrained DINO) | attack success rate: 19.36 -> 21.99 | Hurt |
| add the BMP branch (ablation) | ZePAD · ANIMALS10 (CIFAR10-pretrained DINO) | robust accuracy: 72.47 -> 74.71 | Improved |
| apply ZePAD to the SSL encoder instead of plain baseline | DINO encoder · ANIMALS10 (CIFAR10-pretrained DINO) | robust accuracy: 35.46 -> 74.71 | Improved |
| use ZePAD instead of baseline, CIFAR10-pretrained, SVHN | SSL encoders avg · SVHN (CIFAR10-pretrained) | robust accuracy: 13.11 -> 86.97 | Improved |
| use ZePAD instead of baseline, CIFAR10-pretrained, ANIMALS10 | SSL encoders avg · ANIMALS10 (CIFAR10-pretrained) | benign accuracy: 88.52 -> 98.14 | Improved |
| use ZePAD instead of baseline, CIFAR10-pretrained, STL10 | SSL encoders avg · STL10 (CIFAR10-pretrained) | benign accuracy: 82.63 -> 82.51 | No clear effect |
| use ZePAD instead of baseline, CIFAR10-pretrained, CIFAR10 | SSL encoders avg · CIFAR10 (CIFAR10-pretrained) | benign accuracy: 93.34 -> 93.57 | No clear effect |
| use ZePAD vs Gen-AF defense | SimCLR · STL10 (ImageNet-pretrained) | benign accuracy: 68.69 -> 82.07 | Improved |
| use ZePAD vs Gen-AF defense | SimCLR · STL10 (ImageNet-pretrained) | robust accuracy under AdvEncoder: 62.28 -> 73.26 | Improved |
Closest papers by meaning
- Certified Adversarial Robustness with Additive Noise (NeurIPS 2019 · similarity 0.89)
- Adversarial Robustness of Supervised Sparse Coding (NeurIPS 2020 · similarity 0.88)
- Towards Deep Learning Models Resistant to Adversarial Attacks (ICLR 2018 · similarity 0.88)
- Adversarial Training Should Be Cast as a Non-Zero-Sum Game (ICLR 2024 · similarity 0.87)
- Enhancing Adversarial Transferability with Checkpoints of a Single Model's Training (CVPR 2025 · similarity 0.87)
- Efficient Robustness Evaluation via Constraint Relaxation (AAAI 2025 · similarity 0.87)
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.
