AI Paper Atlas

Research map / Adversarial robustness and security / Certified and empirical robustness

Debiased Dual-Invariant Defense for Adversarially Robust Person Re-Identification

Yuhang Zhou, Yanxiang Zhao, Zhongyun Hua, Zhipu Liu, Zhaoquan Gu, Qing Liao, Leo Yu Zhang

AAAI 2026 · Adversarial robustness and security › Certified and empirical robustness · approach: Attack evaluation and defense critique · 1 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 diffusion-based data balancing (ablation)Metric AT · Market-1501 FNA 8/255-16 ablationmAP: 28.38 -> 29.85Improved
add adversarial feature learning (ablation)Metric AT · Market-1501 FNA 8/255-16 ablationmAP: 28.38 -> 30.34Improved
add self-meta learning (ablation)Metric AT · Market-1501 FNA 8/255-16 ablationmAP: 28.38 -> 29.48Improved
add FNES (farthest negative extension softening) (ablation)Metric AT · Market-1501 FNA 8/255-16 ablationmAP: 28.38 -> 30.98Improved
combine all modules (ablation)Metric AT · Market-1501 FNA 8/255-16 ablationmAP: 28.38 -> 31.99Improved
combine all modules (ablation)Metric AT · DukeMTMC FNA 8/255-16 ablationmAP: 23.76 -> 26.02Improved
add diffusion, adversarial and self-meta modules individually (ablation)Metric AT · DukeMTMC white-box SMA 8/255-16mAP: 40.40 -> 39.44, 38.67, 38.65Hurt
use metric adversarial training instead of PGD classification AT (ablation)AT_PGD · Market-1501 FNA 8/255-16mAP: 26.18 -> 28.38Improved
full method instead of Metric AT on unseen domain (ablation)Metric AT · Market-1501 to DukeMTMC cross-domain FNA 8/255-16mAP: 4.47 -> 6.17Improved
fine-tune only classifier on clean data after adversarial trainingResNet18 AT · CIFAR10 PGDrobust accuracy: 54.045 -> 50.752Hurt
fine-tune only classifier on clean data after adversarial trainingResNet50 AT · CIFAR10 PGDrobust accuracy: 53.076 -> 50.218Hurt
apply the full defense instead of standard adversarial trainingAdv_train · Market-1501 cleanmAP: 69.69 -> 68.50Hurt

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