AI Paper Atlas

Research map / Adversarial robustness and security / Backdoor and poisoning

Pre-trained Trojan Attacks for Visual Recognition

Aishan Liu, Xianglong Liu, Xinwei Zhang, Yisong Xiao, Yuguang Zhou, Siyuan Liang, Jiakai Wang, Xiaochun Cao, Dacheng Tao

IJCV 2025 · Adversarial robustness and security › Backdoor and poisoning · approach: Trigger-based backdoor attacks · 41 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 Pre-trained Trojan instead of classical backdoorsbackdoor attack · ImageNet detectionASR (%): 1.78 -> 88.20Improved
use Pre-trained Trojan instead of BadEncoder on unsupervised encoderSimCLR encoder · ImageNet detectionASR (%): 0.40 -> 37.60Improved
use Pre-trained Trojan instead of BadEncoder on unsupervised encoderSimCLR encoder · ImageNet classificationASR (%): 73.71 -> 98.38Improved
increase number of reference style images from 5 to 40 (ablation)Pre-trained Trojan trigger · ImageNet detectionASR (%): 14.80 -> 91.40Improved
increase style images from 40 to 60 (ablation)Pre-trained Trojan trigger · ImageNet detectionASR (%): 91.40 -> 88.98No clear effect
enlarge trigger patch from 40x40 to 80x80 (ablation)Pre-trained Trojan trigger · ImageNet detectionASR (%): 42.10 -> 88.20Improved
enlarge trigger patch from 80x80 to 100x100 (ablation)Pre-trained Trojan trigger · ImageNet detectionASR (%): 88.20 -> 89.84No clear effect
set style loss weight alpha to 10^5 instead of 10^4 (ablation)Pre-trained Trojan trigger · ImageNet detectionASR (%): 34.62 -> 88.42Improved
set style loss weight alpha to 10^5 instead of 10^6 (ablation)Pre-trained Trojan trigger · ImageNet detectionASR (%): 23.91 -> 88.42Improved
raise poison ratio of Blended from 0.1% to 10%Blended · ImageNet detectionASR (%): 0.03 -> 9.37Improved
attack large ViTAEv2-B by fine-tuningPre-trained Trojan · ImageNet detectionASR (%): 0.70 -> 52.90Improved
infect SMOKE DLA34 backbonePre-trained Trojan · KITTI 3D detectionAP11 easy: 15.56 -> 12.93Hurt

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