AutoRAN 是一种自动化的安全推理劫持方法，利用对齐程度较低（次要）的辅助模型模拟推理轨迹、生成叙事性提示，并迭代优化这些提示，以绕过现代大型推理模型（LRMs）中的安全推理。
⚠️ 免责声明：本仓库仅用于受控安全研究和 AI 安全红队测试。
AutoRAN 遵循三阶段流程：
日志和结果可在 /records 目录中找到。例如：
ls -lh records/
# Clone the repository
git clone {THIS_REPO}
cd AutoRAN
# Install dependencies
pip install -r requirements.txt
# Apply for model access (see HuggingFace link)
# https://huggingface.co/huihui-ai/Qwen3-8B-abliterated/tree/main
# Start the model server (recommended: use tmux or screen)
vllm serve huihui-ai/Qwen3-8B-abliterated --tensor-parallel-size 4 --port 8000
# Edit the attack prompt in demo.py as needed
python demo.py
# Follow the command line instructions.
# You may need to copy questions to GPT-o3, GPT-o4 Mini, or Gemini 2.5-Flash/Pro,
# then paste the results back into the terminal as prompted.
# Clone the repository
git clone {THIS_REPO}
cd AutoRAN
# Install dependencies
pip install -r requirements.txt
# Set up chat2api for automatic ChatGPT interaction:
# https://github.com/lanqian528/chat2api
# Apply for model access (see HuggingFace link)
# https://huggingface.co/huihui-ai/Qwen3-8B-abliterated/tree/main
# Start the model server (recommended: use tmux or screen)
vllm serve huihui-ai/Qwen3-8B-abliterated --tensor-parallel-size 4 --port 8000
# Run the main experiment script
python main.py
此代码仅用于研究和教育目的发布，旨在支持对 LLM 安全漏洞的负责任评估。请勿使用此代码针对真实系统发起攻击，或在受控环境之外生成有害输出。
@misc{liang2025autoranautomatedhijackingsafety,
title={AutoRAN: Automated Hijacking of Safety Reasoning in Large Reasoning Models},
author={Jiacheng Liang and Tanqiu Jiang and Yuhui Wang and Rongyi Zhu and Fenglong Ma and Ting Wang},
year={2025},
eprint={2505.10846},
archivePrefix={arXiv},
primaryClass={cs.LG},
url={https://arxiv.org/abs/2505.10846},
}