AI Paper Atlas

Research map / Adversarial robustness and security / Backdoor and poisoning

Invisible Backdoor Attack With Sample-Specific Triggers

Yuezun Li, Yiming Li, Baoyuan Wu, Longkang Li, Ran He, Siwei Lyu

ICCV 2021 · Adversarial robustness and security › Backdoor and poisoning · approach: Trigger-based backdoor attacks · 516 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 trigger generated from a different test image (inconsistent trigger) (ablation)SSBA · ImageNetASR (%): 99.5 -> 23.3Hurt
use trigger generated from a different test image (inconsistent trigger) (ablation)SSBA · MS-Celeb-1MASR (%): 100 -> 98.1No clear effect
train encoder on another dataset (MS-Celeb-1M for ImageNet) (ablation)SSBA · ImageNetASR (%): 99.5 -> 99.4No clear effect
train encoder on ImageNet for MS-Celeb-1M classifier (ablation)SSBA · MS-Celeb-1MASR (%): 100 -> 99.5No clear effect
change target label to 1, 2 or 3 (ablation)SSBA · ImageNetBA/ASR (%): 85.4/99.4, 85.6/99.3, 85.6/99.5No clear effect
use sample-specific triggers instead of BadNets patchSSBA vs BadNets · ImageNetPSNR: 25.635 -> 27.195Improved
use sample-specific triggers instead of Blended AttackSSBA vs Blended Attack · ImageNetASR (%): 95.8 -> 99.5Improved
use sample-specific triggers instead of Blended AttackSSBA vs Blended Attack · ImageNetPSNR: 45.809 -> 27.195Hurt
increase the poisoning rate gamma (ablation)SSBA · ImageNet and MS-Celeb-1MASR (%): ASR > 95% with only 2% poisoned samplesImproved
prune 20% of neurons (fine-pruning defense)BadNets and Blended Attack vs SSBA · ImageNet and MS-Celeb-1MASR (%): Blended Attack ASR below 10%Improved
replace ImageNet test images by COCO or random noise inputs (ablation)SSBA · ImageNetASR (%): 99.5 -> 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