Saved in:
| Main Authors: | , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2507.22063 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866913965265649664 |
|---|---|
| author | Mo, Wenjie Jacky Liu, Qin Wen, Xiaofei Jung, Dongwon Askari, Hadi Zhou, Wenxuan Zhao, Zhe Chen, Muhao |
| author_facet | Mo, Wenjie Jacky Liu, Qin Wen, Xiaofei Jung, Dongwon Askari, Hadi Zhou, Wenxuan Zhao, Zhe Chen, Muhao |
| contents | Large Language Models (LLMs) for code generation (i.e., Code LLMs) have demonstrated impressive capabilities in AI-assisted software development and testing. However, recent studies have shown that these models are prone to generating vulnerable or even malicious code under adversarial settings. Existing red-teaming approaches rely on extensive human effort, limiting their scalability and practicality, and generally overlook the interactive nature of real-world AI-assisted programming, which often unfolds over multiple turns. To bridge these gaps, we present RedCoder, a red-teaming agent that engages victim models in multi-turn conversation to elicit vulnerable code. The pipeline to construct RedCoder begins with a multi-agent gaming process that simulates adversarial interactions, yielding a set of prototype conversations and an arsenal of reusable attack strategies. We then fine-tune an LLM on these prototype conversations to serve as the backbone of RedCoder. Once deployed, RedCoder autonomously engages Code LLMs in multi-turn conversations, dynamically retrieving relevant strategies from the arsenal to steer the dialogue toward vulnerability-inducing outputs. Experiments across multiple Code LLMs show that our approach outperforms prior single-turn and multi-turn red-team methods in inducing vulnerabilities in code generation, offering a scalable and effective tool for evaluating the security boundaries of modern code-generation systems. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2507_22063 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | RedCoder: Automated Multi-Turn Red Teaming for Code LLMs Mo, Wenjie Jacky Liu, Qin Wen, Xiaofei Jung, Dongwon Askari, Hadi Zhou, Wenxuan Zhao, Zhe Chen, Muhao Software Engineering Artificial Intelligence Large Language Models (LLMs) for code generation (i.e., Code LLMs) have demonstrated impressive capabilities in AI-assisted software development and testing. However, recent studies have shown that these models are prone to generating vulnerable or even malicious code under adversarial settings. Existing red-teaming approaches rely on extensive human effort, limiting their scalability and practicality, and generally overlook the interactive nature of real-world AI-assisted programming, which often unfolds over multiple turns. To bridge these gaps, we present RedCoder, a red-teaming agent that engages victim models in multi-turn conversation to elicit vulnerable code. The pipeline to construct RedCoder begins with a multi-agent gaming process that simulates adversarial interactions, yielding a set of prototype conversations and an arsenal of reusable attack strategies. We then fine-tune an LLM on these prototype conversations to serve as the backbone of RedCoder. Once deployed, RedCoder autonomously engages Code LLMs in multi-turn conversations, dynamically retrieving relevant strategies from the arsenal to steer the dialogue toward vulnerability-inducing outputs. Experiments across multiple Code LLMs show that our approach outperforms prior single-turn and multi-turn red-team methods in inducing vulnerabilities in code generation, offering a scalable and effective tool for evaluating the security boundaries of modern code-generation systems. |
| title | RedCoder: Automated Multi-Turn Red Teaming for Code LLMs |
| topic | Software Engineering Artificial Intelligence |
| url | https://arxiv.org/abs/2507.22063 |