paper-with-me

홈 › Papers

An LLM-Assisted Easy-to-Trigger Backdoor Attack on Code Completion Models: Injecting Disguised Vulnerabilities against Strong Detection

2024-06-10 · Shenao Yan, Shen Wang, Yue Duan, Hanbin Hong, Kiho Lee, Doowon Kim, Yuan Hong

Large Language Models (LLMs) have transformed code completion tasks, providing context-based suggestions to boost developer productivity in software engineering. As users often fine-tune these models for specific applications, poisoning and backdoor attacks can covertly alter the model outputs. To address this critical security challenge, we introduce CodeBreaker, a pioneering LLM-assisted backdoor attack framework on code completion models. Unlike recent attacks that embed malicious payloads in detectable or irrelevant sections of the code (e.g., comments), CodeBreaker leverages LLMs (e.g., GPT-4) for sophisticated payload transformation (without affecting functionalities), ensuring that both the poisoned data for fine-tuning and generated code can evade strong vulnerability detection. CodeBreaker stands out with its comprehensive coverage of vulnerabilities, making it the first to provide such an extensive set for evaluation. Our extensive experimental evaluations and user studies underline the strong attack performance of CodeBreaker across various settings, validating its superiority over existing approaches. By integrating malicious payloads directly into the source code with minimal transformation, CodeBreaker challenges current security measures, underscoring the critical need for more robust defenses for code completion.

📄 PDF Abstract BibTeX arXiv:2406.06822

Code (1)

datasec-lab/codebreaker 공식 구현 jax

Tasks

Backdoor AttackCode CompletionVulnerability Detection

Methods 이 논문이 사용한 방법론

SET Dynamic Sparse Training method where weight mask is updated randomly periodically

Similar Papers 제목 키워드 기반

Imperceptible and Multi-channel Backdoor Attack against Deep Neural Networks

2022-01-31 · Mingfu Xue, Shifeng Ni, Yinghao Wu, Yushu Zhang 외

Recent researches demonstrate that Deep Neural Networks (DNN) models are vulnerable to backdoor attacks. The backdoored DNN model will behave maliciously when images containing backdoor triggers arrive. To date, existing…

Backdoor Attackbackdoor defense

Natural Backdoor Attacks on Speech Recognition Models

2026-07-17 · Jinwen Xin, Xixiang Lyu, Jing Ma arxiv

With the rapid development of deep learning, its vulnerability has gradually emerged in recent years. This work focuses on backdoor attacks on speech recognition systems. We adopt sounds that are ordinary in nature or in…

Speech Recognition

Do Backdoors Assist Membership Inference Attacks?

2023-03-22 · Yumeki Goto, Nami Ashizawa, Toshiki Shibahara, Naoto Yanai

When an adversary provides poison samples to a machine learning model, privacy leakage, such as membership inference attacks that infer whether a sample was included in the training of the model, becomes effective by mov…

Inference AttackMembership Inference Attack

CL-Attack: Textual Backdoor Attacks via Cross-Lingual Triggers

2024-12-26 · Jingyi Zheng, Tianyi Hu, Tianshuo Cong, Xinlei He

Backdoor attacks significantly compromise the security of large language models by triggering them to output specific and controlled content. Currently, triggers for textual backdoor attacks fall into two categories: fix…

Backdoor AttackSentence

UIBDiffusion: Universal Imperceptible Backdoor Attack for Diffusion Models

2024-12-16 · CVPR 2025 1 · Yuning Han, Bingyin Zhao, Rui Chu, Feng Luo 외

Recent studies show that diffusion models (DMs) are vulnerable to backdoor attacks. Existing backdoor attacks impose unconcealed triggers (e.g., a gray box and eyeglasses) that contain evident patterns, rendering remarka…

Backdoor Attack