AI Paper Atlas

Research map / Adversarial robustness and security / Backdoor and poisoning

Seal Your Backdoor with Variational Defense

Ivan Sabolić, Matej Grcić, Siniša Šegvić

ICCV 2025 · Adversarial robustness and security › Backdoor and poisoning · approach: Trigger-based backdoor attacks · 0 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 heuristic data splitting with latent clean-label posterior recovery (VIBE-SS-F)backdoor defense on CIFAR-10 · CIFAR-10 (avg over 8 attacks)ASR (%): 16.0 -> 3.2 vs best baseline ABLImproved
use VIBE-SS on CIFAR-100 versus best baselinebackdoor defense on CIFAR-100 · CIFAR-100 (avg over 4 attacks)ASR (%): 15.5 -> 0.5 vs ASDImproved
add off-manifold community preprocessing (ablation)VIBE on LC clean-label attack · CIFAR-10 LC attackASR (%): 14% -> 6.3%Improved
add self-supervised pre-training on the poisoned data before EM (ablation)VIBE · CIFAR-10ELBO / convergenceImproved
use full corrupted posterior p(y|x,l) instead of approximation p(y|l) (ablation)VIBE-SS · CIFAR-10 (avg over 8 attacks)ASR (%): 6.8 -> 3.2Improved
use frozen DINOv2 foundation-model features instead of self-supervised ResNet-18VIBE · CIFAR-100 transferACC (%): 74.3 -> 92.3Improved
lower the poisoning rate (ablation)VIBE vs other defenses · BadNets CIFAR-10ASRNo clear effect
replace SimCLR pre-training by SimCLR + MIMIC robust pre-training (ablation)VIBE-SS · CIFAR-10 CTRL attackASR (%): 26.8 -> 4.2Improved
apply VIBE to dataset poisoned with both BadNets and LCVIBE-SS-A vs DBD-SS · CIFAR-10 BadNets+LCASR (%): 99.8 -> 1.3Improved
use VIBE instead of baselines on Adap-Blend attackVIBE-SS-F · CIFAR-10 Adap-BlendASR (%): 93.9 -> 9.0 vs ASD; VIBE-SS-A 36.7Mixed

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