视觉大模型 / VLM

A Cross-Modal Prompt Injection Attack against Large Vision-Language Models with Image-Only Perturbation

针对大视觉-语言模型的跨模态提示注入攻击：仅图像扰动

Hao Yang, Zhuo Ma, Yang Liu, Yilong Yang, Guancheng Wang, JianFeng Ma

机构 * Xidian University（西安电子科技大学） 纠错

专题命中 其他VLM ：vision-language model(title,abstract)；分类 cs.CV

AI总结 本文提出CrossMPI攻击，通过仅图像扰动实现跨模态提示注入，改进模型隐藏状态空间优化并采用层选择策略与距离递减扰动策略，有效提升攻击性能。

SkyLink: A Large Vision-Language Model Driven Re-ranking Framework for Cross-View UAV geolocalization

SkyLink：一种由大型视觉-语言模型驱动的跨视图无人机地理定位重排序框架

Bowen Liu, Pengyue Jia, Wanyu Wang, Derong Xu, Jiawei Cheng, Jiancheng Dong, Xiao Han, Zimo Zhao, Chao Zhang, Bowen Yu, Fangyu Hong, Xiangyu Zhao

机构 * Department of Data Science, City University of Hong Kong, Hong Kong（香港城市大学数据科学系） ； Information Systems, City University of Hong Kong, Hong Kong（香港城市大学信息系统系） ； College of Computer Science and Technology, Zhejiang University of Technology, Zhejiang（浙江工业大学计算机科学与技术学院） 纠错

专题命中 其他VLM ：vision-language model(title,abstract)；分类 cs.CV

AI总结 本文提出SkyLink框架，利用大型视觉-语言模型建模无人机与卫星视图间的视觉语义关系，通过引入关系感知损失提升跨视图地理定位的排序效果。

On the Trustworthiness of Generative Foundation Models: Guideline, Assessment, and Perspective

生成基础模型的可信度：指南、评估与视角

Yue Huang, Chujie Gao, Siyuan Wu, Haoran Wang, Xiangqi Wang, Yujun Zhou, Yanbo Wang, Jiayi Ye, Jiawen Shi, Qihui Zhang, Yuan Li, Han Bao, Zhaoyi Liu, Tianrui Guan, Dongping Chen, Ruoxi Chen, Kehan Guo, Andy Zou, Bryan Hooi Kuen-Yew, Caiming Xiong, Elias Stengel-Eskin, Hongyang Zhang, Hongzhi Yin, Huan Zhang, Huaxiu Yao, Jaehong Yoon, Jieyu Zhang, Kai Shu, Kaijie Zhu, Ranjay Krishna, Swabha Swayamdipta, Taiwei Shi, Weijia Shi, Xiang Li, Yiwei Li, Yuexing Hao, Zhihao Jia, Zhize Li, Xiuying Chen, Zhengzhong Tu, Xiyang Hu, Tianyi Zhou, Jieyu Zhao, Lichao Sun, Furong Huang, Or Cohen Sasson, Prasanna Sattigeri, Anka Reuel, Max Lamparth, Yue Zhao, Nouha Dziri, Yu Su, Huan Sun, Heng Ji, Chaowei Xiao, Mohit Bansal, Nitesh V. Chawla, Jian Pei, Jianfeng Gao, Michael Backes, Philip S. Yu, Neil Zhenqiang Gong, Pin-Yu Chen, Bo Li, Dawn Song, Xiangliang Zhang

专题命中 其他VLM ：vision-language model(abstract)

AI总结 本文提出生成基础模型的可信度框架，通过多学科协作制定指导原则，引入动态评估平台TrustGen，揭示可信度进展与挑战，并探讨未来研究方向。