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

Binary Code Patching: An Ancient Art Refined for the...

Dr. Barton P. Miller (Vilas Distinguished Achievement Professor at The University of Wisconsin-Madison)

Read More

SigmaDiff: Semantics-Aware Deep Graph Matching for Pseudocode Diffing

Lian Gao (University of California Riverside), Yu Qu (University of California Riverside), Sheng Yu (University of California, Riverside & Deepbits Technology Inc.), Yue Duan (Singapore Management University), Heng Yin (University of California, Riverside & Deepbits Technology Inc.)

Read More

AVMON: Securing Autonomous Vehicles by Learning Control Invariants and...

Ahmed Abdo, Sakib Md Bin Malek, Xuanpeng Zhao, Nael Abu-Ghazaleh (University of California, Riverside)

Read More