AI Paper Atlas

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

Black-Box Dissector: Towards Erasing-Based Hard-Label Model Stealing Attack

Yixu Wang, Jie Li, Hong Liu, Yan Wang, Yongjian Wu, Feiyue Huang, Rongrong Ji

ECCV 2022 · Adversarial robustness and security › Black-box and transfer attacks · approach: Data-free substitute training · 41 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 CAM-driven erasing (ablation)black-box dissector · CIFAR10, 30k queriestest accuracy: 76.12% -> 80.47%Improved
add self-KD module (ablation)black-box dissector · CIFAR10, 30k queriestest accuracy: 79.02% -> 80.47%Improved
use CAM-prior erasing instead of random erasing (CutOut) (ablation)black-box dissector · CIFAR10test accuracy: 77.11% -> 80.47%Improved
use self-KD instead of CutOut (ablation)black-box dissector · CIFAR10test accuracy: 78.91% -> 80.47%Improved
use self-KD instead of label smoothing (0.9) (ablation)black-box dissector · CIFAR10test accuracy: 78.22% -> 80.47%Improved
use self-KD instead of label smoothing (0.8) (ablation)black-box dissector · CIFAR10test accuracy: 77.46% -> 80.47%Improved
double weight decay on full method (ablation)black-box dissector · CIFAR10test accuracy: 80.47% -> 81.65%Mixed
add self-KD under adaptive misinformation (threshold 0.9) (ablation)black-box dissector · CIFAR10test accuracy: 61.81% -> 74.40%Improved
add self-KD under prediction poisoning (threshold 0.8) (ablation)black-box dissector · CIFAR10test accuracy: 76.49% -> 79.23%Improved
add k-Center selection on small-class datasetsblack-box dissector · CIFAR10test accuracy: 80.47 -> 79.27Hurt
add k-Center selection on many-class datasetblack-box dissector · CUBS200test accuracy: 36.28 -> 44.43Improved
use black-box dissector instead of KnockoffNetshard-label stealing · CIFAR10, transfer attack ResNet-34attack success rate: 57.85% -> 76.63%Improved

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