JailGuard: A Universal Detection Framework for LLM Prompt-based Attacks
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2023
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866929760263733248 |
|---|---|
| author | Zhang, Xiaoyu Zhang, Cen Li, Tianlin Huang, Yihao Jia, Xiaojun Hu, Ming Zhang, Jie Liu, Yang Ma, Shiqing Shen, Chao |
| author_facet | Zhang, Xiaoyu Zhang, Cen Li, Tianlin Huang, Yihao Jia, Xiaojun Hu, Ming Zhang, Jie Liu, Yang Ma, Shiqing Shen, Chao |
| contents | The systems and software powered by Large Language Models (LLMs) and Multi-Modal LLMs (MLLMs) have played a critical role in numerous scenarios. However, current LLM systems are vulnerable to prompt-based attacks, with jailbreaking attacks enabling the LLM system to generate harmful content, while hijacking attacks manipulate the LLM system to perform attacker-desired tasks, underscoring the necessity for detection tools. Unfortunately, existing detecting approaches are usually tailored to specific attacks, resulting in poor generalization in detecting various attacks across different modalities. To address it, we propose JailGuard, a universal detection framework deployed on top of LLM systems for prompt-based attacks across text and image modalities. JailGuard operates on the principle that attacks are inherently less robust than benign ones. Specifically, JailGuard mutates untrusted inputs to generate variants and leverages the discrepancy of the variants' responses on the target model to distinguish attack samples from benign samples. We implement 18 mutators for text and image inputs and design a mutator combination policy to further improve detection generalization. The evaluation on the dataset containing 15 known attack types suggests that JailGuard achieves the best detection accuracy of 86.14%/82.90% on text and image inputs, outperforming state-of-the-art methods by 11.81%-25.73% and 12.20%-21.40%. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2312_10766 |
| institution | arXiv |
| publishDate | 2023 |
| record_format | arxiv |
| spellingShingle | JailGuard: A Universal Detection Framework for LLM Prompt-based Attacks Zhang, Xiaoyu Zhang, Cen Li, Tianlin Huang, Yihao Jia, Xiaojun Hu, Ming Zhang, Jie Liu, Yang Ma, Shiqing Shen, Chao Cryptography and Security The systems and software powered by Large Language Models (LLMs) and Multi-Modal LLMs (MLLMs) have played a critical role in numerous scenarios. However, current LLM systems are vulnerable to prompt-based attacks, with jailbreaking attacks enabling the LLM system to generate harmful content, while hijacking attacks manipulate the LLM system to perform attacker-desired tasks, underscoring the necessity for detection tools. Unfortunately, existing detecting approaches are usually tailored to specific attacks, resulting in poor generalization in detecting various attacks across different modalities. To address it, we propose JailGuard, a universal detection framework deployed on top of LLM systems for prompt-based attacks across text and image modalities. JailGuard operates on the principle that attacks are inherently less robust than benign ones. Specifically, JailGuard mutates untrusted inputs to generate variants and leverages the discrepancy of the variants' responses on the target model to distinguish attack samples from benign samples. We implement 18 mutators for text and image inputs and design a mutator combination policy to further improve detection generalization. The evaluation on the dataset containing 15 known attack types suggests that JailGuard achieves the best detection accuracy of 86.14%/82.90% on text and image inputs, outperforming state-of-the-art methods by 11.81%-25.73% and 12.20%-21.40%. |
| title | JailGuard: A Universal Detection Framework for LLM Prompt-based Attacks |
| topic | Cryptography and Security |
| url | https://arxiv.org/abs/2312.10766 |