6 papers · 1 filter
Rethinking Backdoor Repair Evaluation: Distinguishing Aggregate Clean Utility from Benign Performance Preservation
Baogang Song, Changtian Song, Jian Chen +4
Backdoor repair aims to suppress malicious behavior in compromised models while preserving benign task performance. Existing studies typically evaluate these objectives using Attac…
Benign on Label, Malicious by Design: Clean-Label Dormant-to-Activated Backdoor via Machine Unlearning with Removable Camouflage
Dongdong Zhao, Can Li, Xiang Yao +3
Existing backdoor attacks often become effective immediately after backdoor implantation and may therefore be exposed before exploitation. Machine unlearning activated dormant back…
From Classification to Consistent Templates: Multiple Permuted-Label Classifier Encoding for Biometric Template Protection
Baogang Song, Zhongshu Zhao, Qianrong Zheng +2
Biometric template protection (BTP) must secure stored templates while tolerating intra-class variations. Existing methods rely on protected-domain similarity matching, error corre…
Discard the Dross and Select the Essential: Pre-query Sample Selection for Black-box Membership Inference Attacks
Dongdong Zhao, Jinrong Hu, Changtian Song +3
Black-box membership inference attacks (MIAs) rely on target-model queries to infer whether candidate samples were used for training. However, membership signals are highly non-uni…
Injection, Attack and Erasure: Revocable Backdoor Attacks via Machine Unlearning
Baogang Song, Dongdong Zhao, Jianwen Xiang +2
Backdoor attacks pose a persistent security risk to deep neural networks (DNNs) due to their stealth and durability. While recent research has explored leveraging model unlearning…
BioDeepHash: Mapping Biometrics into a Stable Code
Baogang Song, Dongdong Zhao, Jiang Yan +2
With the wide application of biometrics, more and more attention has been paid to the security of biometric templates. However most of existing biometric template protection (BTP)…