MazeBreaker: Multi-Agent Reinforcement Learning for Dynamic Jailbreaking of LLM Security Defenses
Open Access
- 12 April 2026
- conference paper
- Published by Association for Computing Machinery (ACM)
- p. 3028-3039
Abstract
No abstract availableThis publication has 9 references indexed in Scilit:
- FuzzLLM: A Novel and Universal Fuzzing Framework for Proactively Discovering Jailbreak Vulnerabilities in Large Language Models2024
- A survey on large language model (LLM) security and privacy: The Good, The Bad, and The UglyHigh-Confidence Computing, 2024
- When LLM Meets DRL: Advancing Jailbreaking Efficiency via DRL-guided Search2024
- Reinforcement Learning-Driven LLM Agent for Automated Attacks on LLMs2024
- A Wolf in Sheep’s Clothing: Generalized Nested Jailbreak Prompts can Fool Large Language Models Easily2024
- MASTERKEY: Automated Jailbreaking of Large Language Model Chatbots2024
- Jailbroken: How Does LLM Safety Training Fail?2023
- Red Teaming Language Models with Language Models2022
- A contextual-bandit approach to personalized news article recommendation2010