arXiv · Computer Vision· Jingwen Huang, Yiming Fang, Hanxiao Wu, Boyuan He, Keyi Luo·· 4 hr agoAI score29
TerraAgent:面向长上下文遥感视觉问答的基于演示的多智能体推理框架
TerraAgent: Demonstration-Grounded Multi-Agent Reasoning for Long-Context Remote Sensing Visual Question Answering
AI brief
论文提出 AVA 框架,以免训练方式包装冻结的遥感多模态大语言模型,包含 GDR、MACE、CVA 三个模块。
Source: arXiv · Computer Vision · arxiv.org