AI Paper Atlas

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

Membership Inference Attacks for Unseen Classes

Pratiksha Thaker, Neil Kale, Steven Wu, Virginia Smith

ICML 2026 · Adversarial robustness and security › Attacks on text and graphs · approach: Membership inference and inversion · 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
exclude the target class from shadow model training (unseen-class setting) (ablation)LiRA shadow-model attack · CINIC-10 and CIFAR-100 (superclasses)TPR at 1% FPRHurt
use quantile regression instead of LiRA shadow models on unseen CIFAR-100 classesMembership inference attack · CIFAR-100 superclasses, unseen classTPR at 1% FPR: 1.4% -> 3.8%Improved
use quantile regression instead of shadow models on CINIC-10Membership inference attack · CINIC-10, unseen classTPR at 1% FPRImproved
use RMIA instead of LiRAShadow-model attack in unseen-class setting · CIFAR-100 / CINIC-10 unseen classTPR at low FPRHurt
use top-two logit difference as score function instead of true-label confidence (ablation)Quantile regression attack · Class dropoutTPR at 1% FPRImproved
use top-two logit difference as score function instead of true-label confidence (ablation)Shadow-model attack · Class dropoutTPR at 1% FPRHurt
early stopping of the quantile regression model (ablation)Quantile regression attack · CINIC-10 / CIFAR-100attack performanceNo clear effect
leave out 100 to 900 of 1000 classes from attack training (ablation)Quantile regression attack · ImageNet, unseen classesTPR at 1% FPR: stays around 3.8% (0.4% at 0.1% FPR)No clear effect
reduce attack training data to 10 samples per class (ablation)Quantile regression attack · ImageNetTPR at 1% FPR: 9.0% -> 3.9%Hurt
use quantile regression instead of RMIA on tabular and text dataMembership inference attack · Texas and 20 NewsgroupsTPR at 1% FPR: up to 2x on Texas, up to 6x on 20 NewsgroupsImproved
increase regularization defense (weight decay, differential privacy) (ablation)Target model · Unseen-class MIAMIA successHurt

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