AI Paper Atlas

Research map / Adversarial robustness and security / Black-box and transfer attacks

On the Adversarial Transferability of Generalized “Skip Connections”

Yisen Wang, Yichuan Mo, Dongxian Wu, Mingjie Li, Xingjun Ma, Zhouchen Lin

TPAMI 2026 · Adversarial robustness and security › Black-box and transfer attacks · approach: Transferable adversarial examples

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 changedWhereMeasuredEffect
add SGM to PGDPGD ResNet-50 source · VGG16 targetattack success rate: 45.00 -> 58.68Improved
add SGM to SIASIA ResNet-50 source · VGG16 targetattack success rate: 90.88 -> 95.93Improved
add SGM to PGD with ViT-B sourcePGD ViT-B source · VGG16 targetattack success rate: 16.34% to 21.04%Improved
add SGM to PGD with Inception-V3 source (no skip connections)PGD Inception-V3 source · VGG16 targetattack success rate: 29.74 -> 45.43Improved
add SGM to MFAA in model ensembleMFAA ensemble source · SE154 targetattack success rate: 61.70% to 90.32%Improved
add SGM to Logit targeted attackLogit ResNet-50 source · VGG16 targetattack success rate: 37.90 -> 63.49Improved
add SGM to RFA against CFA defenseRFA · CFA defenseattack success rate: 9.40% to 13.13%Improved
add MUP baseline to PGD with Inception-V3 sourcePGD Inception-V3 source · VGG16 targetattack success rate: 29.74 -> 21.62Hurt
add MUP baseline to PGD with ResNet-50 sourcePGD ResNet-50 source · VGG16 targetattack success rate: 45.00 -> 50.84Improved
decay gradient of residual modules (lower gamma) for ResNet-50 source (ablation)SGM PGD ResNet-50 · ImageNet black-boxattack success rate: peak around gamma = 0.2Improved
decay gradients for Inception-V3 source (no skip connections) (ablation)SGM PGD Inception-V3 · ImageNet black-boxattack success rate: optimal around gamma = 0.6Improved
lower gamma from 1.0 to 0.5 on adversarial confidence on true class (ablation)PGD+SGM image (a) · ImageNet samplesground-truth confidence (lower is better): 16.84 -> 0.56Improved

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