AI Paper Atlas

Research map / Adversarial robustness and security / Backdoor and poisoning

Better Trigger Inversion Optimization in Backdoor Scanning

Guanhong Tao, Guangyu Shen, Yingqi Liu, Shengwei An, Qiuling Xu, Shiqing Ma, Pan Li, Xiangyu Zhang

CVPR 2022 · Adversarial robustness and security › Backdoor and poisoning · approach: Trigger-based backdoor attacks · 54 citations (OpenAlex)

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
Replace NC's mask-and-perturbation optimization with direct per-pixel perturbation optimization using two tanh functionsImageNet model trigger inversion · ImageNet validation, loggerhead turtle to kangarooASR: 28.00% -> 70.00%Improved
Replace NC's mask-and-perturbation optimization with direct per-pixel perturbation optimizationImageNet model trigger inversion · ImageNet validation, loggerhead turtle to kangarooperturbed pixels: 26583 -> 822Improved
Use direct per-pixel perturbation optimization instead of NCImageNet model trigger inversion · ImageNet validation, loggerhead turtle to kangarootime (minutes): 9.02 -> 4.27Improved
Replace the full tanh pixel-change formulation in Equation 4 with the clipped linear formulation in Equation 6 (ablation)Ours on clean models generating natural triggers · Natural trigger generationtrigger qualityNo clear effect
Use the full tanh pixel-change formulation rather than the clipped linear formulation (ablation)Ours on trojaned models during backdoor scanning · Backdoor scanningtrigger generation qualityImproved
Retain the tanh functions in the pixel-count penalty Lpixel (ablation)Ours · Trigger generationtrigger qualityImproved
Use Ours instead of NC for trigger inversionImageNet model trigger inversion · ImageNet validation, loggerhead turtle to snowbirdASR: 60.00% -> 72.00%Improved
Use Ours instead of NC for trigger inversionCIFAR-10 clean ResNet20 · CIFAR-10, all class pairsASR after matching trigger size: 39.83% -> 78.22%Improved
Use Ours instead of NC to harden the modelResNet32 · SVHN model hardeningaverage trigger-size increase: 152.02%Improved
Replace the NC-based optimization component in K-arm with OursK-arm polygon-backdoor scanner · TrojAI competition modelsdetection accuracy: 2%Improved
Use Ours instead of NC for inversion of WaNet backdoorsBackdoor scanning · WaNetanomaly indexImproved

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