AI Paper Atlas

Research map / Adversarial robustness and security / Attacks on text and graphs

AdvDiff: Generating Unrestricted Adversarial Examples using Diffusion Models

Xuelong Dai, Kaisheng Liang, Bin Xiao

ECCV 2024 · Adversarial robustness and security › Attacks on text and graphs · approach: Diffusion-based adversarial purification · 63 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
use diffusion-based AdvDiff instead of GAN-based U-GAN on MNIST LeNet5 against PGD-ATunrestricted adversarial attack · MNIST LeNet5 (PGD-AT)ASR (%): 79.4 -> 88.6Improved
replace PGD-gradient guidance with AdvDiff adversarial guidance in diffusion attacksdiffusion adversarial sampling · ImageNet generation qualityFID: 20.5 -> 16.2 (DiffAttack vs AdvDiff)Improved
AdvDiff instead of U-BigGAN on generation qualityunrestricted adversarial attack · ImageNet generation qualityIS: 129.4 -> 343.8Improved
AdvDiff instead of DiffAttack/AdvDiffuser on generation timediffusion attacks · ImageNet ResNet50Time (s): 28.2 -> 9.2 (vs DiffAttack)Improved
use targeted AdvDiff for black-box transfer vs earlier attacksAdvDiff · ImageNet transfer to ResNet-152ASR (%): 32.5 -> 20.5 (AutoAttack vs AdvDiff)Hurt
use untargeted AdvDiff for transfer and defensesAdvDiff · ImageNet against DiffPureASR (%): 41.6 -> 75.2Improved
increase noise sampling guidance steps N (ablation)AdvDiff · ImageNet ResNet50ASR vs generation qualityMixed
set adversarial guidance scale extremely large (ablation)AdvDiff · ImageNet ResNet50image realismHurt
apply adversarial guidance in early reverse timesteps (ablation)AdvDiff · ImageNet ResNet50ASRNo clear effect

Later papers that compared with this method

2 later accepted papers report AdvDiff as a baseline in their results tables; 2 of them report a better number on most of the results they share with it.

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