AI Paper Atlas

Research map / Label-efficient and robust learning / Self-supervised and contrastive

SSL-Cleanse: Trojan Detection and Mitigation in Self-Supervised Learning

Mengxin Zheng, Jiaqi Xue, Zihao Wang, Xun Chen, Qian Lou, Lei Jiang, Xiaofeng Wang

ECCV 2024 · Label-efficient and robust learning › Self-supervised and contrastive · approach: Data-efficient and robust self-supervision · 26 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
replace Kneedle with Sliding Window Kneedle for cluster number estimation (ablation)SSL-Cleanse detector · ImageNet-100DACC: predicted K 20 -> 100; 30% DACC improvement (80%)Improved
increase fraction of unlabeled data used for detection from 5% to 10% (ablation)SSL-Cleanse detector · CIFAR-10 (CTRL, BYOL)DACC: 71 -> 80Improved
increase fraction of unlabeled data used for detection from 5% to 10% (ablation)SSL-Cleanse detector · ImageNet-100 (CTRL, BYOL)DACC: 82 -> 85Improved
apply SCU mitigation to backdoored encodersTrojaned SSL encoders · CIFAR-10 SSL-Backdoor BYOLASR (%): 48.32 -> 1.14Improved
apply SCU mitigation to backdoored encodersTrojaned SSL encoders · CIFAR-10 CTRL SimCLRASR (%): 81.84 -> 1.15Improved
apply SCU mitigation to backdoored encodersTrojaned SSL encoders · ImageNet-100 SSL-Backdoor SimCLRclean ACC (%): 60.18 -> 58.58Hurt
apply SCU mitigation to encoders for other downstream tasksTrojaned SSL encoders · Downstream STL-10ASR (%): 36.55 -> 1.84Improved
apply SCU mitigation to encoders for other downstream tasksTrojaned SSL encoders · Downstream GTSRBASR (%): 38.28 -> 1.03Improved
use NC or ABS linear-probe detection when downstream data differs from encoder dataNC and ABS detectors · SSL-Backdoor, STL-10 downstreamNC Anomaly Index / ABS REASR: 2.05 -> 1.42 (NC); 0.89 -> 0.34 (ABS)Hurt

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