With the rising popularity of Large Language Models (LLMs), assessing their
trustworthiness through security tasks has gained critical importance.
Regarding the new task of universal goal hijacking, previous efforts have
concentrated solely on optimization algorithms, overlooking the crucial role of
the prompt. To fill this gap, we propose a universal goal hijacking method
called POUGH that incorporates semantic-guided prompt processing strategies.
Specifically, the method starts with a sampling strategy to select
representative prompts from a candidate pool, followed by a ranking strategy
that prioritizes the prompts. Once the prompts are organized sequentially, the
method employs an iterative optimization algorithm to generate the universal
fixed suffix for the prompts. Experiments conducted on four popular LLMs and
ten types of target responses verified the effectiveness of our method.

通过语义导向的提示处理策略，我们提出了一种称为 POUGH 的通用目标劫持方法，用于评估大型语言模型的可信度，并在四个流行的语言模型和十种目标响应类型上进行了实验证实其有效性。