学会观看却忘记遵循：视觉指令微调使 LLM 更易遭受越狱攻击
Learning To See But Forgetting To Follow: Visual Instruction Tuning Makes LLMs More Prone To Jailbreak Attacks
AI总结：
研究对比三种最先进 VLM 及其 LLM 主干，发现视觉指令微调会削弱安全护栏、使模型更易被越狱，并提出安全评估与训练建议。
摘要：
为大型语言模型（LLM）增强图像理解能力，催生了大量高性能视觉语言模型（VLM）。尽管研究 LLM 与人类价值观的对齐已受到广泛关注，但 VLM 的安全性尚未获得同等重视。本文探讨了越狱对三种最先进 VLM 的影响，这三种模型分别采用不同的建模方法。通过将每种 VLM 与其各自的 LLM 主干进行比较，我们发现每种 VLM 都更容易受到越狱攻击。我们将此视为视觉指令微调带来的不良后果：它会对 LLM 的安全护栏产生遗忘效应。因此，我们基于旨在揭示 VLM 弱点的评估策略，为未来工作提出建议，并主张在视觉指令微调过程中将安全措施纳入考量。
英文摘要：
Augmenting Large Language Models (LLMs) with image-understanding capabilities has resulted in a boom of high-performing Vision-Language models (VLMs). While studying the alignment of LLMs to human values has received widespread attention, the safety of VLMs has not received the same attention. In this paper, we explore the impact of jailbreaking on three state-of-the-art VLMs, each using a distinct modeling approach. By comparing each VLM to their respective LLM backbone, we find that each VLM is more susceptible to jailbreaking. We consider this as an undesirable outcome from visual instruction-tuning, which imposes a forgetting effect on an LLM's safety guardrails. Therefore, we provide recommendations for future work based on evaluation strategies that aim to highlight the weaknesses of a VLM, as well as take safety measures into account during visual instruction tuning.
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。
