arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2406.12259cs.AI
收藏

针对医学领域大型语言模型的对抗性攻击

Adversarial Attacks on Large Language Models in Medicine

National Library of Medicine (NLM), National Institutes of Health (NIH)(美国国立卫生研究院国家医学图书馆);University of Maryland at College Park(马里兰大学帕克分校)

Yifan Yang, Qiao Jin, Furong Huang, Zhiyong Lu

AI总结:

本研究探讨医学领域大型语言模型对对抗性攻击的脆弱性,发现开源和专有模型在多个医学任务中易受操纵,且领域特定任务需要更多对抗性数据,同时攻击会导致模型权重变化,为防御提供途径。

摘要:

大型语言模型(LLMs)在医疗应用中的整合为医学诊断、治疗建议和患者护理方面带来了有前景的进展。然而,LLMs对对抗性攻击的敏感性构成了重大威胁,可能在敏感的医疗环境中导致有害结果。本研究调查了LLMs在三个医学任务中对两种类型对抗性攻击的脆弱性。利用真实世界的患者数据,我们证明了开源和专有LLMs在多个任务中都容易受到操纵。这项研究进一步揭示,领域特定任务在模型微调中需要比通用领域任务更多的对抗性数据才能有效执行攻击,尤其是对于能力更强的模型。我们发现,虽然整合对抗性数据并未显著降低模型在医学基准上的整体性能,但它确实导致微调模型权重发生明显变化,这为检测和对抗模型攻击提供了一条潜在途径。这项研究强调了在医疗应用中保护LLMs的迫切需求,需要采取强有力的安全措施和开发防御机制,以确保它们在医疗环境中的安全有效部署。

英文摘要:

The integration of Large Language Models (LLMs) into healthcare applications offers promising advancements in medical diagnostics, treatment recommendations, and patient care. However, the susceptibility of LLMs to adversarial attacks poses a significant threat, potentially leading to harmful outcomes in delicate medical contexts. This study investigates the vulnerability of LLMs to two types of adversarial attacks in three medical tasks. Utilizing real-world patient data, we demonstrate that both open-source and proprietary LLMs are susceptible to manipulation across multiple tasks. This research further reveals that domain-specific tasks demand more adversarial data in model fine-tuning than general domain tasks for effective attack execution, especially for more capable models. We discover that while integrating adversarial data does not markedly degrade overall model performance on medical benchmarks, it does lead to noticeable shifts in fine-tuned model weights, suggesting a potential pathway for detecting and countering model attacks. This research highlights the urgent need for robust security measures and the development of defensive mechanisms to safeguard LLMs in medical applications, to ensure their safe and effective deployment in healthcare settings.

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。

↑