AI Paper Atlas

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

Improving Adversarial Transferability via Intermediate-level Perturbation Decay

Qizhang Li, Yiwen Guo, Wangmeng Zuo, Hao Chen

NeurIPS 2023 · Adversarial robustness and security › Black-box and transfer attacks · approach: Transferable adversarial examples · 3 citations (OpenAlex)

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 ILPD to I-FGSMI-FGSM · ImageNetaverage attack success rate: 22.36 -> 57.06Improved
add ILPD to I-FGSMI-FGSM · CIFAR-10average attack success rate: 48.62 -> 62.82Improved
use ILPD instead of second-best ILA++I-FGSM · ImageNetaverage attack success rate: +10.07Improved
add ILPD to MI-FGSMMI-FGSM · ImageNetaverage attack success rate: 28.57 -> 59.84Improved
add ILPD to DI$^2$-FGSMDI$^2$-FGSM · ImageNetaverage attack success rate: 46.88 -> 63.23Improved
add ILPD to AdmixAdmix · ImageNetaverage attack success rate: 35.59 -> 60.28Improved
add ILPD, attack robust modelsI-FGSM · ImageNet robust victimsaverage attack success rate: 10.56 -> 24.66Improved
add ILPD vs NAA, attack smoothed ResNet-50I-FGSM · random smoothing ResNet-50attack success rate: 7.20 -> 12.90Improved
add ILPD, attack NRP-purified imagesI-FGSM · NRP defenseaverage attack success rate: 9.39 -> 20.03Improved
use ILPD with ViT-B substitute instead of NAAILA++/NAA · ImageNetaverage attack success rate: 43.52 -> 46.63Improved
use ILPD with Swin-B substitute instead of NAAILA++/NAA · ImageNetaverage attack success rate: 21.41 -> 36.77Improved
choose the split position at the first layer (ablation)ILPD · ImageNetaverage attack success rate: 53.12No clear effect

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.

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

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