Compromesso！意大利语多示例越狱攻击削弱大语言模型的安全性
Compromesso! Italian Many-Shot Jailbreaks Undermine the Safety of Large Language Models
AI总结：
本文研究意大利语多示例越狱攻击对LLM安全性的影响，构建意大利语不安全问答数据集，发现四个开源LLM家族在少量不安全示例下即出现不安全行为，且随示例增多迅速恶化。
摘要：
随着不同语言社区和用户采用大语言模型（LLMs），评估其跨语言安全性变得至关重要。尽管人们持续努力使LLMs变得安全，它们仍然可能通过越狱攻击被诱导做出不安全行为，越狱攻击是一种促使模型在其操作准则之外行事的技术。然而，关于LLM安全性和越狱攻击的研究迄今主要集中在英语上，限制了我们对LLMs在其他语言中安全性的理解。我们通过研究多示例越狱攻击在意大利语中的有效性来缩小这一差距，在该攻击中，模型被给予不安全示例以诱导不安全行为。为了支持我们的分析，我们创建了一个新的意大利语不安全问答对数据集。利用该数据集，我们在四个开源权重LLM家族中识别出明确的安全漏洞。我们发现，即使仅给出少量不安全示例，这些模型也会表现出不安全行为，并且——更令人担忧的是——这种倾向会随着示例数量的增加而迅速升级。
英文摘要：
As diverse linguistic communities and users adopt large language models (LLMs), assessing their safety across languages becomes critical. Despite ongoing efforts to make LLMs safe, they can still be made to behave unsafely with jailbreaking, a technique in which models are prompted to act outside their operational guidelines. Research on LLM safety and jailbreaking, however, has so far mostly focused on English, limiting our understanding of LLM safety in other languages. We contribute towards closing this gap by investigating the effectiveness of many-shot jailbreaking, where models are prompted with unsafe demonstrations to induce unsafe behaviour, in Italian. To enable our analysis, we create a new dataset of unsafe Italian question-answer pairs. With this dataset, we identify clear safety vulnerabilities in four families of open-weight LLMs. We find that the models exhibit unsafe behaviors even when prompted with few unsafe demonstrations, and -- more alarmingly -- that this tendency rapidly escalates with more demonstrations.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
