Subtoxic Questions: Dive Into Attitude Change of LLM's Response in Jailbreak Attempts
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_ | 1866909167413886976 |
|---|---|
| author | Zhang, Tianyu Zhao, Zixuan Huang, Jiaqi Hua, Jingyu Zhong, Sheng |
| author_facet | Zhang, Tianyu Zhao, Zixuan Huang, Jiaqi Hua, Jingyu Zhong, Sheng |
| contents | As Large Language Models (LLMs) of Prompt Jailbreaking are getting more and more attention, it is of great significance to raise a generalized research paradigm to evaluate attack strengths and a basic model to conduct subtler experiments. In this paper, we propose a novel approach by focusing on a set of target questions that are inherently more sensitive to jailbreak prompts, aiming to circumvent the limitations posed by enhanced LLM security. Through designing and analyzing these sensitive questions, this paper reveals a more effective method of identifying vulnerabilities in LLMs, thereby contributing to the advancement of LLM security. This research not only challenges existing jailbreaking methodologies but also fortifies LLMs against potential exploits. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2404_08309 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Subtoxic Questions: Dive Into Attitude Change of LLM's Response in Jailbreak Attempts Zhang, Tianyu Zhao, Zixuan Huang, Jiaqi Hua, Jingyu Zhong, Sheng Cryptography and Security Artificial Intelligence Computation and Language As Large Language Models (LLMs) of Prompt Jailbreaking are getting more and more attention, it is of great significance to raise a generalized research paradigm to evaluate attack strengths and a basic model to conduct subtler experiments. In this paper, we propose a novel approach by focusing on a set of target questions that are inherently more sensitive to jailbreak prompts, aiming to circumvent the limitations posed by enhanced LLM security. Through designing and analyzing these sensitive questions, this paper reveals a more effective method of identifying vulnerabilities in LLMs, thereby contributing to the advancement of LLM security. This research not only challenges existing jailbreaking methodologies but also fortifies LLMs against potential exploits. |
| title | Subtoxic Questions: Dive Into Attitude Change of LLM's Response in Jailbreak Attempts |
| topic | Cryptography and Security Artificial Intelligence Computation and Language |
| url | https://arxiv.org/abs/2404.08309 |