Report
Graph of Attacks with Pruning: Optimizing Stealthy Jailbreak Prompt Generation for Enhanced LLM Content Moderation
| العنوان: | Graph of Attacks with Pruning: Optimizing Stealthy Jailbreak Prompt Generation for Enhanced LLM Content Moderation |
|---|---|
| المؤلفون: | Schwartz, Daniel, Bespalov, Dmitriy, Wang, Zhe, Kulkarni, Ninad, Qi, Yanjun |
| سنة النشر: | 2025 |
| المجموعة: | Computer Science |
| مصطلحات موضوعية: | Computer Science - Cryptography and Security, Computer Science - Artificial Intelligence, Computer Science - Computation and Language |
| الوصف: | We present a modular pipeline that automates the generation of stealthy jailbreak prompts derived from high-level content policies, enhancing LLM content moderation. First, we address query inefficiency and jailbreak strength by developing Graph of Attacks with Pruning (GAP), a method that utilizes strategies from prior jailbreaks, resulting in 92% attack success rate on GPT-3.5 using only 54% of the queries of the prior algorithm. Second, we address the cold-start issue by automatically generating seed prompts from the high-level policy using LLMs. Finally, we demonstrate the utility of these generated jailbreak prompts of improving content moderation by fine-tuning PromptGuard, a model trained to detect jailbreaks, increasing its accuracy on the Toxic-Chat dataset from 5.1% to 93.89%. Comment: 15 pages, 7 figures |
| نوع الوثيقة: | Working Paper |
| URL الوصول: | http://arxiv.org/abs/2501.18638 |
| رقم الانضمام: | edsarx.2501.18638 |
| قاعدة البيانات: | arXiv |
| الوصف غير متاح. |