LLMs易受伪装成科学语言的恶意提示攻击
LLMs are Vulnerable to Malicious Prompts Disguised as Scientific Language
AI总结：
本文发现多款主流LLMs会被伪装成社会科学证据的恶意提示诱导，产生更高偏见与毒性，并分析作者、发表场域和对话进程等影响因素。
摘要：
随着大型语言模型（LLMs）被部署到各种真实场景中，人们对其可能传播危害的担忧日益增加。已有多种越狱技术被开发出来，用以揭示这些模型的脆弱性并提升其安全性。本研究揭示，许多最先进的LLMs容易受到隐藏在科学语言背后的恶意请求影响。具体而言，我们在GPT4o、GPT4o-mini、GPT-4、LLama3-405B-Instruct、Llama3-70B-Instruct、Cohere和Gemini模型上的实验表明，当提示请求故意曲解社会科学和心理学研究，将其作为支持刻板印象偏见益处的证据时，模型的偏见和毒性会显著增加。令人警惕的是，这些模型还可能被操纵生成声称偏见有益的捏造科学论证，恶意行为者可利用这些论证系统性地越狱这些强大的LLMs。我们的分析研究了导致模型易受学术语言恶意请求影响的多种因素。提及作者姓名和发表场域会增强模型的说服力，并且偏见分数会随着对话推进而上升。我们的发现呼吁更审慎地研究科学数据在训练LLMs中的使用。
英文摘要：
As large language models (LLMs) have been deployed in various real-world settings, concerns about the harm they may propagate have grown. Various jailbreaking techniques have been developed to expose the vulnerabilities of these models and improve their safety. This work reveals that many state-of-the-art LLMs are vulnerable to malicious requests hidden behind scientific language. Specifically, our experiments with GPT4o, GPT4o-mini, GPT-4, LLama3-405B-Instruct, Llama3-70B-Instruct, Cohere, Gemini models demonstrate that, the models' biases and toxicity substantially increase when prompted with requests that deliberately misinterpret social science and psychological studies as evidence supporting the benefits of stereotypical biases. Alarmingly, these models can also be manipulated to generate fabricated scientific arguments claiming that biases are beneficial, which can be used by ill-intended actors to systematically jailbreak these strong LLMs. Our analysis studies various factors that contribute to the models' vulnerabilities to malicious requests in academic language. Mentioning author names and venues enhances the persuasiveness of models, and the bias scores increase as dialogues progress. Our findings call for a more careful investigation on the use of scientific data for training LLMs.
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。
