AI Paper Atlas

Research map / Adversarial robustness and security / Certified and empirical robustness

Adversarial Training for Probabilistic Robustness

Yi Zhang, Yuhang Chen, Zhen Chen, Wenjie Ruan, Xiaowei Huang, Siddartha Khastgir, Xingyu Zhao

ICCV 2025 · Adversarial robustness and security › Certified and empirical robustness · approach: Adversarial training refinements · 4 citations (OpenAlex)

Read the paper

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 changedWhereMeasuredEffect
train with AT-PR instead of AT-WCR (PGD)ResNet-18 · CIFAR-100 PR(eps=0.15)PR: 80.47 -> 87.38Improved
train with AT-PR instead of AT-WCR (PGD)ResNet-18 · CIFAR-10 PR(eps=0.15)PR: 89.78 -> 93.21Improved
train with AT-PR instead of AT-WCR (PGD)ResNet-18 · SVHN PR(eps=0.15)PR: 95.41 -> 97.84Improved
train with AT-PR instead of AT-WCR (PGD)WRN-50-2 · SVHN PR(eps=0.15)PR: 96.77 -> 96.26Hurt
train with AT-PR instead of AT-WCR (PGD)ResNet-18 · CIFAR-10 PGD accuracyPGD accuracy: 45.01 -> 39.77Hurt
train with AT-PR instead of AT-WCR (PGD)ResNet-34 · Tiny-ImageNet clean accuracyaccuracy: 49.33 -> 53.98Improved
replace PGD AE with farthest-from-boundary AE (AT-PR) vs random AE from Step 1 (ablation)AT-PR · CIFAR-10 PR(eps=0.15)PR: 92.64 -> 93.21Improved
replace PGD AE with farthest-from-boundary AE (AT-PR) vs random AE from Step 1 (ablation)AT-PR · CIFAR-100 PR(eps=0.15)PR: 85.21 -> 87.38Improved
use AT-PR instead of PGD for lower train-test PR gapResNet-18 · CIFAR-100 PR gap (eps=0.15)generalization gap: 8.54 -> 2.87Improved
use AT-PR instead of PGD for lower train-test PR gapResNet-18 · Tiny-ImageNet PR gap (eps=0.15)generalization gap: 11.94 -> 7.66Improved
add AT-PR fine-tuning to TRADESViT-B · CIFAR-10clean accuracy: 83.55 -> 86.03Improved
add AT-PR fine-tuning to TRADESViT-B · CIFAR-10PGD-20 accuracy: 53.14 -> 52.94Hurt

Closest papers by meaning

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.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy