|国家预印本平台
首页|Automatic LLM Red Teaming

Automatic LLM Red Teaming

Automatic LLM Red Teaming

来源:Arxiv_logoArxiv
英文摘要

Red teaming is critical for identifying vulnerabilities and building trust in current LLMs. However, current automated methods for Large Language Models (LLMs) rely on brittle prompt templates or single-turn attacks, failing to capture the complex, interactive nature of real-world adversarial dialogues. We propose a novel paradigm: training an AI to strategically `break' another AI. By formalizing red teaming as a Markov Decision Process (MDP) and employing a hierarchical Reinforcement Learning (RL) framework, we effectively address the inherent sparse reward and long-horizon challenges. Our generative agent learns coherent, multi-turn attack strategies through a fine-grained, token-level harm reward, enabling it to uncover subtle vulnerabilities missed by existing baselines. This approach sets a new state-of-the-art, fundamentally reframing LLM red teaming as a dynamic, trajectory-based process (rather than a one-step test) essential for robust AI deployment.

Roman Belaire、Arunesh Sinha、Pradeep Varakantham

计算技术、计算机技术

Roman Belaire,Arunesh Sinha,Pradeep Varakantham.Automatic LLM Red Teaming[EB/OL].(2025-08-06)[2025-08-23].https://arxiv.org/abs/2508.04451.点此复制

评论