Research map / Adversarial robustness and security / Certified and empirical robustness
CIARD: Cyclic Iterative Adversarial Robustness Distillation
Liming Lu, Shuchao Pang, Xu Zheng, Xiang Gu, Anan Du, Yunhuai Liu, Yongbin Zhou
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 Iterative Teacher Training (ablation) | CIARD (ResNet-18) · CIFAR-10 | PGD-TRADES robust accuracy: 53.84 -> 54.03 | Improved |
| add Iterative Teacher Training (ablation) | CIARD (ResNet-18) · CIFAR-10 | clean accuracy: 87.76 -> 88.63 | Improved |
| add contrastive push loss on top of ITT (ablation) | CIARD (ResNet-18) · CIFAR-10 | PGD-TRADES robust accuracy: 54.03 -> 54.54 | Improved |
| add contrastive push loss on top of ITT (ablation) | CIARD (ResNet-18) · CIFAR-10 | clean accuracy: 88.63 -> 88.86 | Improved |
| add contrastive push loss on top of ITT (ablation) | CIARD (ResNet-18) · CIFAR-10 | weighted accuracy: 71.30 -> 71.70 | Improved |
| CIARD vs B-MTARD dual-teacher method | ResNet-18 · CIFAR-10 FGSM | weighted robustness: 74.81 -> 75.38 | Improved |
| CIARD vs B-MTARD dual-teacher method | ResNet-18 · CIFAR-100 PGD-SAT | robust accuracy: 28.50 -> 28.05 | Hurt |
| CIARD vs B-MTARD dual-teacher method | MobileNet-V2 · CIFAR-100 FGSM | weighted robustness: 50.25 -> 50.14 | Hurt |
| CIARD vs B-MTARD dual-teacher method | MobileNet-V2 · CIFAR-10 FGSM | clean accuracy: 89.09 -> 89.51 | Improved |
| CIARD vs B-MTARD dual-teacher method | ResNet-18 · CIFAR-10 Square Attack | weighted robustness: 84.01 -> 84.45 | Improved |
Later papers that compared with this method
1 later accepted paper reports CIARD as a baseline in their results tables; 1 of them report a better number on most of the results they share with it.
- GCMRD: Global Consistency Multi-teacher Robustness Distillation (ECCV 2026 · better on 11 of 11 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
- GCMRD: Global Consistency Multi-teacher Robustness Distillation (ECCV 2026 · similarity 0.92)
- Out of Thin Air: Exploring Data-Free Adversarial Robustness Distillation (AAAI 2024 · similarity 0.90)
- Indirect Gradient Matching for Adversarial Robust Distillation (ICLR 2025 · similarity 0.89)
- Improving Adversarial Robust Fairness via Anti-Bias Soft Label Distillation (NeurIPS 2024 · similarity 0.88)
- Robust Distillation via Untargeted and Targeted Intermediate Adversarial Samples (CVPR 2024 · similarity 0.88)
- Learning Better Certified Models from Empirically-Robust Teachers (NeurIPS 2026 · similarity 0.88)
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.
