1
Harnessing Textual Refusal Directions for Multimodal Safety
利用文本拒绝方向增强多模态AI安全性的前沿研究，为模型防御提供新思路。
arXiv:2606.31876v1 Announce Type: new Abstract: To improve safety in Large Language Models (LLMs) we can either perform post-training alignment or exp…