Jailbreak

暑期文献阅读一：A Comprehensive Study of Jailbreak Attack versus Defense for Large Language Models

九种攻击技术和七种防御机制。这些方法应用于三个不同的语言模型：Vicuna、LLaMA和GPT-3.5 Turbo。使用RoBERTa模型进行微调，分类恶意响应Jailbreak Attack T

暑期文献 Study Comprehensive Jailbreak

admin 6月前

103 0