Making Them Ask and Answer: Jailbreaking Large Language Models in Few Queries via Disguise and Reconstruction
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2024
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866911911441858560 |
|---|---|
| author | Liu, Tong Zhang, Yingjie Zhao, Zhe Dong, Yinpeng Meng, Guozhu Chen, Kai |
| author_facet | Liu, Tong Zhang, Yingjie Zhao, Zhe Dong, Yinpeng Meng, Guozhu Chen, Kai |
| contents | In recent years, large language models (LLMs) have demonstrated notable success across various tasks, but the trustworthiness of LLMs is still an open problem. One specific threat is the potential to generate toxic or harmful responses. Attackers can craft adversarial prompts that induce harmful responses from LLMs. In this work, we pioneer a theoretical foundation in LLMs security by identifying bias vulnerabilities within the safety fine-tuning and design a black-box jailbreak method named DRA (Disguise and Reconstruction Attack), which conceals harmful instructions through disguise and prompts the model to reconstruct the original harmful instruction within its completion. We evaluate DRA across various open-source and closed-source models, showcasing state-of-the-art jailbreak success rates and attack efficiency. Notably, DRA boasts a 91.1% attack success rate on OpenAI GPT-4 chatbot. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2402_18104 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Making Them Ask and Answer: Jailbreaking Large Language Models in Few Queries via Disguise and Reconstruction Liu, Tong Zhang, Yingjie Zhao, Zhe Dong, Yinpeng Meng, Guozhu Chen, Kai Cryptography and Security Artificial Intelligence In recent years, large language models (LLMs) have demonstrated notable success across various tasks, but the trustworthiness of LLMs is still an open problem. One specific threat is the potential to generate toxic or harmful responses. Attackers can craft adversarial prompts that induce harmful responses from LLMs. In this work, we pioneer a theoretical foundation in LLMs security by identifying bias vulnerabilities within the safety fine-tuning and design a black-box jailbreak method named DRA (Disguise and Reconstruction Attack), which conceals harmful instructions through disguise and prompts the model to reconstruct the original harmful instruction within its completion. We evaluate DRA across various open-source and closed-source models, showcasing state-of-the-art jailbreak success rates and attack efficiency. Notably, DRA boasts a 91.1% attack success rate on OpenAI GPT-4 chatbot. |
| title | Making Them Ask and Answer: Jailbreaking Large Language Models in Few Queries via Disguise and Reconstruction |
| topic | Cryptography and Security Artificial Intelligence |
| url | https://arxiv.org/abs/2402.18104 |