Stand on The Shoulders of Giants: Building JailExpert from Previous Attack Experience
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866915465509470208 |
|---|---|
| author | Wang, Xi Jian, Songlei Li, Shasha Li, Xiaopeng Ji, Bin Ma, Jun Liu, Xiaodong Wang, Jing Bao, Feilong Zhang, Jianfeng Wang, Baosheng Yu, Jie |
| author_facet | Wang, Xi Jian, Songlei Li, Shasha Li, Xiaopeng Ji, Bin Ma, Jun Liu, Xiaodong Wang, Jing Bao, Feilong Zhang, Jianfeng Wang, Baosheng Yu, Jie |
| contents | Large language models (LLMs) generate human-aligned content under certain safety constraints. However, the current known technique ``jailbreak prompt'' can circumvent safety-aligned measures and induce LLMs to output malicious content. Research on Jailbreaking can help identify vulnerabilities in LLMs and guide the development of robust security frameworks. To circumvent the issue of attack templates becoming obsolete as models evolve, existing methods adopt iterative mutation and dynamic optimization to facilitate more automated jailbreak attacks. However, these methods face two challenges: inefficiency and repetitive optimization, as they overlook the value of past attack experiences. To better integrate past attack experiences to assist current jailbreak attempts, we propose the \textbf{JailExpert}, an automated jailbreak framework, which is the first to achieve a formal representation of experience structure, group experiences based on semantic drift, and support the dynamic updating of the experience pool. Extensive experiments demonstrate that JailExpert significantly improves both attack effectiveness and efficiency. Compared to the current state-of-the-art black-box jailbreak methods, JailExpert achieves an average increase of 17\% in attack success rate and 2.7 times improvement in attack efficiency. Our implementation is available at \href{https://github.com/xiZAIzai/JailExpert}{XiZaiZai/JailExpert} |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2508_19292 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Stand on The Shoulders of Giants: Building JailExpert from Previous Attack Experience Wang, Xi Jian, Songlei Li, Shasha Li, Xiaopeng Ji, Bin Ma, Jun Liu, Xiaodong Wang, Jing Bao, Feilong Zhang, Jianfeng Wang, Baosheng Yu, Jie Cryptography and Security Artificial Intelligence Large language models (LLMs) generate human-aligned content under certain safety constraints. However, the current known technique ``jailbreak prompt'' can circumvent safety-aligned measures and induce LLMs to output malicious content. Research on Jailbreaking can help identify vulnerabilities in LLMs and guide the development of robust security frameworks. To circumvent the issue of attack templates becoming obsolete as models evolve, existing methods adopt iterative mutation and dynamic optimization to facilitate more automated jailbreak attacks. However, these methods face two challenges: inefficiency and repetitive optimization, as they overlook the value of past attack experiences. To better integrate past attack experiences to assist current jailbreak attempts, we propose the \textbf{JailExpert}, an automated jailbreak framework, which is the first to achieve a formal representation of experience structure, group experiences based on semantic drift, and support the dynamic updating of the experience pool. Extensive experiments demonstrate that JailExpert significantly improves both attack effectiveness and efficiency. Compared to the current state-of-the-art black-box jailbreak methods, JailExpert achieves an average increase of 17\% in attack success rate and 2.7 times improvement in attack efficiency. Our implementation is available at \href{https://github.com/xiZAIzai/JailExpert}{XiZaiZai/JailExpert} |
| title | Stand on The Shoulders of Giants: Building JailExpert from Previous Attack Experience |
| topic | Cryptography and Security Artificial Intelligence |
| url | https://arxiv.org/abs/2508.19292 |