TroubleLLM: Align to Red Team Expert

Author:

Zhuoer Xu, Jianping Zhang, Shiwen Cui, Changhua Meng, Weiqiang Wang

Keyword:

Computer Science, Artificial Intelligence, Artificial Intelligence (cs.AI), Computation and Language (cs.CL)

journal:

date:

2024-02-28 00:00:00

Abstract

Large Language Models (LLMs) become the start-of-the-art solutions for a variety of natural language tasks and are integrated into real-world applications. However, LLMs can be potentially harmful in manifesting undesirable safety issues like social biases and toxic content. It is imperative to assess its safety issues before deployment. However, the quality and diversity of test prompts generated by existing methods are still far from satisfactory. Not only are these methods labor-intensive and require large budget costs, but the controllability of test prompt generation is lacking for the specific testing domain of LLM applications. With the idea of LLM for LLM testing, we propose the first LLM, called TroubleLLM, to generate controllable test prompts on LLM safety issues. Extensive experiments and human evaluation illustrate the superiority of TroubleLLM on generation quality and generation controllability.

PDF: TroubleLLM: Align to Red Team Expert.pdf