AI Paper Atlas

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

Recognizing Object by Components With Human Prior Knowledge Enhances Adversarial Robustness of Deep Neural Networks

Xiao Li, Ziqi Wang, Bo Zhang, Fuchun Sun, Xiaolin Hu

TPAMI 2023 · Adversarial robustness and security › Attacks on text and graphs · approach: Biologically inspired robust vision · 40 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 part segmentation supervision (w/ P, w/o K vs w/o P) (ablation)ROCK · Pascal-Part-C PGD-40 eps=1accuracy: 14.4 -> 19.4Improved
add human prior knowledge linkage rules on top of part segmentation (ablation)ROCK · Pascal-Part-C PGD-40 eps=1accuracy: 19.4 -> 27.5Improved
add importance weights of linkage rules (ablation)ROCK · Pascal-Part-C PGD-40 eps=1accuracy: 27.5 -> 28.2Improved
add human prior knowledge (benign accuracy) (ablation)ROCK · Pascal-Part-C benignaccuracy: 84.8 -> 80.7Hurt
replace ROW-50 with ROCK under white-box PGD-40ROW-50 · Pascal-Part-C eps=1accuracy: 3.5 -> 28.2Improved
replace ROW_X-101 with ROCK under SPSA black-boxROW_X-101 · Pascal-Part-C eps=8accuracy: 9.8 -> 52.3Improved
combine with TRADES-AWP-EMA adversarial trainingROCK vs ROW-50 · PartImageNet-C AutoAttack eps=8accuracy: 11.5 -> 17.3Improved
use ResNeXt-101 backbone for ROCK with TRADES-AWP-EMAROCK · Pascal-Part-C PGD-40 eps=8accuracy: 32.6 -> 34.5Improved
use DeiT-Ti transformer backbone in ROCKDeiT-Ti · Pascal-Part-C PGD-40 eps=1accuracy: 15.8 -> 27.0Improved
use only 10 part-labeled images per category instead of all part labels (ablation)ROCK · PartImageNet-C 50 categories PGD-40 eps=1accuracy: 19.4 -> 14.5Hurt
replace real part labels with unsupervised pseudo part labels (ablation)ROCK · PartImageNet-C 45 categories PGD-40 eps=1accuracy: 22.6 -> 19.6Hurt
use ROCK instead of ROW-50 on common corruptionsROW-50 · PartImageNet-C corruptions standard trainingmCA: 35.8 -> 40.4Improved

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