arXiv preprint 2025 / 高效动作解码

CEED-VLA: Consistency Vision-Language-Action Model with Early-Exit Decoding

作者：Wenxuan Song, Jiayi Chen, Pengxiang Ding, Yuxin Huang, Han Zhao, Donglin Wang, Haoang Li · 单位：HKUST(GZ), Westlake University, Zhejiang University · 会议/期刊：arXiv preprint 2025 · 方向：高效动作解码 · 发布日期：2025-06-16

视觉语言动作高效动作解码高效推理令牌优化
新标签打开 AlphaXiv新标签打开 Project新标签打开 Code
CEED-VLA: Consistency Vision-Language-Action Model with Early-Exit Decoding figure
AlphaXiv 中文概览（可滚动查看）