AI Paper Atlas

Research map / Adversarial robustness and security / Attacks on text and graphs

When Priors Backfire: On the Vulnerability of Unlearnable Examples to Pretraining

Zhihao Li, Gezheng Xu, Jiale Cai, Ruiyi Fang, Di Wu, Qicheng Lao, Charles Ling, Boyu Wang

ICLR 2026 · Adversarial robustness and security › Attacks on text and graphs · approach: Attacks and defenses in sequential settings · 1 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
add random-class target labels (stage 2) after hard-negative stage (ablation)BAIT · CIFAR-100test accuracy (%): 35.75 -> 23.94Improved
add most-dissimilar-class target labels (stage 3) (ablation)BAIT · CIFAR-100test accuracy (%): 23.94 -> 20.77Improved
add stages 2 and 3 to curriculum on CIFAR-10 (ablation)BAIT · CIFAR-10test accuracy (%): 23.68 -> 14.40Improved
add stages 2 and 3 to curriculum on SVHN (ablation)BAIT · SVHNtest accuracy (%): 21.75 -> 14.37Improved
use ImageNet-pretrained surrogate for REM (ablation)REM · CIFAR-10 (ResNet-18)test accuracy (%): 74.46 -> 56.08Improved
use ImageNet-pretrained surrogate for EMN (ablation)EMN · CIFAR-10 (ResNet-18)test accuracy (%): 61.82 -> 59.84No clear effect
use ImageNet-pretrained surrogate for TUE (ablation)TUE · CIFAR-10 (ResNet-18)test accuracy (%): 82.72 -> 82.65No clear effect
apply Mixup defense (ablation)BAIT · CIFAR-10 (ImageNet-pretrained)test accuracy (%): 14.40 -> 21.08Hurt
apply Cutout defense (ablation)BAIT · CIFAR-10 (ImageNet-pretrained)test accuracy (%): 14.40 -> 13.69No clear effect
apply JPEG compression quality 10 (ablation)BAIT · CIFAR-10 (ImageNet-pretrained)test accuracy (%): 15.91 -> 68.08Hurt
evaluate with randomly initialized backbone instead of pretrainedBAIT · CIFAR-10 ResNet-18test accuracy (%): 14.40 -> 10.18Improved
use larger ViT-based pretrained backbonesUE baselines · CIFAR-10test accuracy (%)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