تقرير
ASTPrompter: Weakly Supervised Automated Language Model Red-Teaming to Identify Likely Toxic Prompts
العنوان: | ASTPrompter: Weakly Supervised Automated Language Model Red-Teaming to Identify Likely Toxic Prompts |
---|---|
المؤلفون: | Hardy, Amelia F., Liu, Houjun, Lange, Bernard, Kochenderfer, Mykel J. |
سنة النشر: | 2024 |
المجموعة: | Computer Science |
مصطلحات موضوعية: | Computer Science - Computation and Language |
الوصف: | Typical schemes for automated red-teaming large language models (LLMs) focus on discovering prompts that trigger a frozen language model (the defender) to generate toxic text. This often results in the prompting model (the adversary) producing text that is unintelligible and unlikely to arise. Here, we propose a reinforcement learning formulation of the LLM red-teaming task which allows us to discover prompts that both (1) trigger toxic outputs from a frozen defender and (2) have low perplexity as scored by the defender. We argue these cases are most pertinent in a red-teaming setting because of their likelihood to arise during normal use of the defender model. We solve this formulation through a novel online and weakly supervised variant of Identity Preference Optimization (IPO) on GPT-2 and GPT-2 XL defenders. We demonstrate that our policy is capable of generating likely prompts that also trigger toxicity. Finally, we qualitatively analyze learned strategies, trade-offs of likelihood and toxicity, and discuss implications. Source code is available for this project at: https://github.com/sisl/ASTPrompter/. Comment: 9 pages, 2 tables, 2 figures |
نوع الوثيقة: | Working Paper |
URL الوصول: | http://arxiv.org/abs/2407.09447 |
رقم الأكسشن: | edsarx.2407.09447 |
قاعدة البيانات: | arXiv |
الوصف غير متاح. |