AI Paper Atlas

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

TkML-AP: Adversarial Attacks to Top-k Multi-Label Learning

Shu Hu, Lipeng Ke, Xin Wang, Siwei Lyu

ICCV 2021 · Adversarial robustness and security › Attacks on text and graphs · approach: Defenses against universal perturbations · 38 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 TkML-AP-U instead of kFooluntargeted attack · MS COCO 2014 top-3ASR (%): 61.4 -> 100Improved
use TkML-AP-U instead of kFooluntargeted attack · PASCAL VOC 2012 top-10ASR (%): 88.7 -> 98.3Improved
use TkML-AP-U instead of kFool (per-pixel perturbation)untargeted attack · MS COCO 2014 top-10Pert (x10^-2): 16.44 -> 0.59Improved
attack with k=3 but system uses larger k'=5TkML-AP-U · PASCAL VOC 2012ASR (%): 99.6 -> 3.6Hurt
attack with k=3 but system uses larger k'=10TkML-AP-U · MS COCO 2014ASR (%): 100 -> 3.9Hurt
use TkML-AP-Uv instead of kUAPsuniversal untargeted attack · MS COCO 2014 top-1ASR (%): 63.9 -> 86.5Improved
use TkML-AP-Uv instead of kUAPs (perturbation)universal untargeted attack · MS COCO 2014 top-1Pert: 0.51 -> 0.13Improved
use TkML-AP-Uv instead of kUAPsuniversal untargeted attack · MS COCO 2014 top-3ASR (%): 73.2 -> 80.5Improved
use TkML-AP-T instead of ML-APtargeted attack Worst case · PASCAL VOC 2012 top-10ASR (%): 39.1 -> 57Improved
use TkML-AP-T instead of ML-APtargeted attack Worst case · MS COCO 2014 top-10ASR (%): 59 -> 73.1Improved
use TkML-AP-T instead of ML-APtargeted attack Random case · PASCAL VOC 2012 top-10ASR (%): 67.7 -> 76.4Improved
use TkML-AP-T instead of ML-APtargeted attack Best case · MS COCO 2014 top-3ASR (%): 100 -> 100No clear effect

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