FDI: Attack Neural Code Generation Systems through User Feedback Channel
This program is tentative and subject to change.
Neural code generation systems have recently attracted increasing attention to improve developer productivity and speed up software development. Typically, these systems maintain a pre-trained neural model and make it available to general users as a service (e.g., through remote APIs) and incorporate a feedback mechanism to extensively collect and utilize the users’ reaction to the generated code, i.e., user feedback. However, the security implications of such feedback have not yet been explored. With a systematic study of current feedback mechanisms, we find that feedback makes these systems vulnerable to \emph{feedback data injection} (FDI) attacks. We discuss the methodology of FDI attacks and present a pre-attack profiling strategy to infer the attack constraints of a targeted system in the black-box setting. We demonstrate two proof-of-concept examples utilizing the FDI attack surface to implement prompt injection attacks and backdoor attacks on practical neural code generation systems. Attackers may stealthily manipulate a neural code generation system to generate code with vulnerabilities, attack payload, and malicious and spam messages. Our findings reveal the security implications of feedback mechanisms in neural code generation systems, paving the way for increasing their security.
This program is tentative and subject to change.
Fri 20 SepDisplayed time zone: Amsterdam, Berlin, Bern, Rome, Stockholm, Vienna change
15:30 - 16:30 | |||
15:30 20mTalk | FDI: Attack Neural Code Generation Systems through User Feedback Channel Technical Papers Zhensu Sun Singapore Management University, Xiaoning Du Monash University, Australia, Xiapu Luo The Hong Kong Polytechnic University, Fu Song State Key Laboratory of Computer Science and Institute of Software, Chinese Academy of Sciences., David Lo Singapore Management University, Li Li Beihang University | ||
15:50 20mTalk | CoderUJB: An Executable and Unified Java Benchmark for Practical Programming Scenarios Technical Papers Zhengran Zeng Peking University, Yidong Wang Peking University, Rui Xie Peking University, Wei Ye Peking University, Shikun Zhang Peking University DOI | ||
16:10 20mTalk | CLAP: Learning Transferable Binary Code Representations with Natural Language Supervision Technical Papers Hao Wang Tsinghua University, Zeyu Gao Tsinghua University, Chao Zhang Tsinghua University, Zihan Sha Information Engineering University, Mingyang Sun University of Electronic Science and Technology of China, Yuchen Zhou Beijing University of Technology, Wenyu Zhu Tsinghua University, Wenju Sun , Han Qiu Tsinghua University, Xi Xiao Tsinghua Shenzhen International Graduate School, Tsinghua University DOI |