Hengzhi Pei (UIUC), Jinyuan Jia (UIUC, Penn State), Wenbo Guo (UC Berkeley, Purdue University), Bo Li (UIUC), Dawn Song (UC Berkeley)

Backdoor attacks have become a major security threat for deploying machine learning models in security-critical applications. Existing research endeavors have proposed many defenses against backdoor attacks. Despite demonstrating certain empirical defense efficacy, none of these techniques could provide a formal and provable security guarantee against arbitrary attacks. As a result, they can be easily broken by strong adaptive attacks, as shown in our evaluation. In this work, we propose TextGuard, the first provable defense against backdoor attacks on text classification. In particular, TextGuard first divides the (backdoored) training data into sub-training sets, achieved by splitting each training sentence into sub-sentences. This partitioning ensures that a majority of the sub-training sets do not contain the backdoor trigger. Subsequently, a base classifier is trained from each sub-training set, and their ensemble provides the final prediction. We theoretically prove that when the length of the backdoor trigger falls within a certain threshold, TextGuard guarantees that its prediction will remain unaffected by the presence of the triggers in training and testing inputs. In our evaluation, we demonstrate the effectiveness of TextGuard on three benchmark text classification tasks, surpassing the certification accuracy of existing certified defenses against backdoor attacks. Furthermore, we propose additional strategies to enhance the empirical performance of TextGuard. Comparisons with state-of-the-art empirical defenses validate the superiority of TextGuard in countering multiple backdoor attacks. Our code and data are available at https://github.com/AI-secure/TextGuard.

View More Papers

Secure Control of Connected and Automated Vehicles Using Trust-Aware...

H M Sabbir Ahmad, Ehsan Sabouni, Akua Dickson (Boston University), Wei Xiao (Massachusetts Institute of Technology), Christos Cassandras, Wenchao Li (Boston University)

Read More

FreqFed: A Frequency Analysis-Based Approach for Mitigating Poisoning Attacks...

Hossein Fereidooni (Technical University of Darmstadt), Alessandro Pegoraro (Technical University of Darmstadt), Phillip Rieger (Technical University of Darmstadt), Alexandra Dmitrienko (University of Wuerzburg), Ahmad-Reza Sadeghi (Technical University of Darmstadt)

Read More

WIP: Hidden Hub Eavesdropping Attack in Matter-enabled Smart Home...

Song Liao, Jingwen Yan, Long Cheng (Clemson University)

Read More

Front-running Attack in Sharded Blockchains and Fair Cross-shard Consensus

Jianting Zhang (Purdue University), Wuhui Chen (Sun Yat-sen University), Sifu Luo (Sun Yat-sen University), Tiantian Gong (Purdue University), Zicong Hong (The Hong Kong Polytechnic University), Aniket Kate (Purdue University)

Read More