AI Paper Atlas

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

BlackboxBench: A Comprehensive Benchmark of Black-Box Adversarial Attacks

Meixi Zheng, Xuanchen Yan, Zihao Zhu, Hongrui Chen, Baoyuan Wu

TPAMI 2025 · Adversarial robustness and security › Black-box and transfer attacks · approach: Query-efficient black-box attacks · 42 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
combine SIA with model-fusion method EnslogitSIA transfer attack · ImageNet untargeted l_infASR (%): 75.1 -> 78.5Improved
combine SIA with CWASIA transfer attack · ImageNet untargeted l_infASR (%): 75.1 -> 77.1Improved
combine individual transfer attacks into pairwise compositestransfer-based attacks · ImageNet untargeted l_inf, 98 compositesASRImproved
combine SIA with model-tuning methods (LGV, Bayesian attack)SIA transfer attack · ImageNet untargeted l_infASRNo clear effect
increase perturbation budget from 1/255 to 32/255 (ablation)Parsimonious Attack · ImageNet score-based untargetedASR: 0.27 -> 1.00Improved
increase perturbation budget from smallest to 32/255 (ablation)Square Attack · ImageNet score-based untargetedASR: 0.31 -> 1.00Improved
increase perturbation budget to 32/255 (ablation)NES · ImageNet score-based untargetedASR: -> 0.79Improved
use surrogate models for gradient estimation (combination methods)score-based attacks (MCG, PRGF, BASES, CG-attack) · ImageNet score-basedASR across budgetsImproved
use semantic-perturbation transfer attacks (FIA, NAA, DRA, IAA) with low norm budget (ablation)transfer-based attacks · ImageNet transfer, varying epsilonASRHurt
add more iterations to momentum-based attacks (MI, NI, PI, CWA) (ablation)transfer-based attacks · ImageNet transferASRHurt
use complex target architectures (ConvNeXt-T, ViT-B/16) instead of ResNet-50 or VGG-19query-based attacks · ImageNet untargeted l_infAQN/ASR ratioImproved

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