arXiv preprint 2025 / 高效注意力

dVLA: Diffusion Vision-Language-Action Model with Multimodal Chain-of-Thought

作者：Junjie Wen, Minjie Zhu, Jiaming Liu, Zhiyuan Liu, Yicun Yang, Linfeng Zhang, Shanghang Zhang, Yichen Zhu, Yi Xu · 单位：Midea Group Peking University, Shanghai Jiaotong University · 会议/期刊：arXiv preprint 2025 · 方向：高效注意力 · 发布日期：2025-09-30

视觉语言动作高效注意力高效推理高效架构扩散策略
新标签打开 AlphaXiv
dVLA: Diffusion Vision-Language-Action Model with Multimodal Chain-of-Thought figure
AlphaXiv 中文概览（可滚动查看）