AutoRAN: Weak-to-Strong Jailbreaking of Large Reasoning Models

2025-05-16 · Jiacheng Liang, Tanqiu Jiang, Yuhui Wang, Rongyi Zhu, Fenglong Ma, Ting Wang

This paper presents AutoRAN, the first automated, weak-to-strong jailbreak attack framework targeting large reasoning models (LRMs). At its core, AutoRAN leverages a weak, less-aligned reasoning model to simulate the target model's high-level reasoning structures, generates narrative prompts, and iteratively refines candidate prompts by incorporating the target model's intermediate reasoning steps. We evaluate AutoRAN against state-of-the-art LRMs including GPT-o3/o4-mini and Gemini-2.5-Flash across multiple benchmark datasets (AdvBench, HarmBench, and StrongReject). Results demonstrate that AutoRAN achieves remarkable success rates (approaching 100%) within one or a few turns across different LRMs, even when judged by a robustly aligned external model. This work reveals that leveraging weak reasoning models can effectively exploit the critical vulnerabilities of much more capable reasoning models, highlighting the need for improved safety measures specifically designed for reasoning-based models. The code for replicating AutoRAN and running records are available at: (https://github.com/JACKPURCELL/AutoRAN-public). (warning: this paper contains potentially harmful content generated by LRMs.)

📄 PDF Abstract BibTeX arXiv:2505.10846

Code (1)

jackpurcell/autoran-public 공식 구현

AutoRank: MCDA Based Rank Personalization for LoRA-Enabled Distributed Learning

2024-12-20 · Shuaijun Chen, Omid Tavallaie, Niousha Nazemi, Xin Chen 외

As data volumes expand rapidly, distributed machine learning has become essential for addressing the growing computational demands of modern AI systems. However, training models in distributed environments is challenging…

Federated Learning

Weak-to-Strong Jailbreaking on Large Language Models

2024-01-30 · Xuandong Zhao, Xianjun Yang, Tianyu Pang, Chao Du 외

Large language models (LLMs) are vulnerable to jailbreak attacks - resulting in harmful, unethical, or biased text generations. However, existing jailbreaking methods are computationally costly. In this paper, we propose…

AutORAN: LLM-driven Natural Language Programming for Agile xApp Development

2026-03-19 · Xin Li, Shiming Yu, Leming Shen, Jianing Zhang 외 arxiv

Traditional RAN systems are closed and monolithic, stifling innovation. The openness and programmability enabled by Open Radio Access Network (O-RAN) are envisioned to revolutionize cellular networks with control-plane a…

Defending Large Language Models Against Jailbreaking Attacks Through Goal Prioritization

2023-11-15 · Zhexin Zhang, Junxiao Yang, Pei Ke, Fei Mi 외

While significant attention has been dedicated to exploiting weaknesses in LLMs through jailbreaking attacks, there remains a paucity of effort in defending against these attacks. We point out a pivotal factor contributi…

Instruction Following

Jailbreaking Large Vision Language Models in Intelligent Transportation Systems

2025-11-17 · Badhan Chandra Das, Md Tasnim Jawad, Md Jueal Mia, M. Hadi Amini 외 arxiv

Large Vision Language Models (LVLMs) demonstrate strong capabilities in multimodal reasoning and many real-world applications, such as visual question answering. However, LVLMs are highly vulnerable to jailbreaking attac…

Visual Question Answering Multimodal Reasoning