Publication Type
Conference Proceeding Article
Version
publishedVersion
Publication Date
7-2023
Abstract
Backdoor attacks for neural code models have gained considerable attention due to the advancement of code intelligence. However, most existing works insert triggers into task-specific data for code-related downstream tasks, thereby limiting the scope of attacks. Moreover, the majority of attacks for pre-trained models are designed for understanding tasks. In this paper, we propose task-agnostic backdoor attacks for code pre-trained models. Our backdoored model is pre-trained with two learning strategies (i.e., Poisoned Seq2Seq learning and token representation learning) to support the multi-target attack of downstream code understanding and generation tasks. During the deployment phase, the implanted backdoors in the victim models can be activated by the designed triggers to achieve the targeted attack. We evaluate our approach on two code understanding tasks and three code generation tasks over seven datasets. Extensive experiments demonstrate that our approach can effectively and stealthily attack code-related downstream tasks.
Keywords
Backdoors, Code understanding, Codegeneration, Deployment phasis, Down-stream, Learning strategy, Multi-targets, Neural code
Discipline
Databases and Information Systems | Information Security
Research Areas
Data Science and Engineering
Publication
Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics, Toronto, Canada, July 9-14
First Page
7236
Last Page
7254
ISBN
9781959429722
Publisher
Association for Computational Linguistics (ACL)
City or Country
Ohio, USA
Citation
LI, Yanzhou; LIU, Shangqing; CHEN, Kangjie; XIE, Xiaofei; ZHANG, Tianwei; and LIU, Yang.
Multi-target backdoor attacks for code pre-trained models. (2023). Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics, Toronto, Canada, July 9-14. 7236-7254.
Available at: https://ink.library.smu.edu.sg/sis_research/8541
Creative Commons License
This work is licensed under a Creative Commons Attribution-NonCommercial-No Derivative Works 4.0 International License.