世界模型

Xiaomi OneVL: One-Step Latent Reasoning and Planning with Vision-Language Explanation

小米OneVL：基于视觉-语言解释的一步潜在推理与规划

Jinghui Lu, Jiayi Guan, Zhijian Huang, Jinlong Li, Guang Li, Lingdong Kong, Yingyan Li, Han Wang, Shaoqing Xu, Yuechen Luo, Fang Li, Chenxu Dang, Junli Wang, Tao Xu, Jing Wu, Jianhua Wu, Xiaoshuai Hao, Wen Zhang, Tianyi Jiang, Lingfeng Zhang, Lei Zhou, Yingbo Tang, Jie Wang, Yinfeng Gao, Xizhou Bu, Haochen Tian, Yihang Qiu, Feiyang Jia, Lin Liu, Yigu Ge, Hanbing Li, Yuannan Shen, Jianwei Cui, Hongwei Xie, Bing Wang, Haiyang Sun, Jingwei Zhao, Jiahui Huang, Pei Liu, Zeyu Zhu, Yuncheng Jiang, Zibin Guo, Chuhong Gong, Hanchao Leng, Kun Ma, Naiyan Wang, Guang Chen, Kuiyuan Yang, Hangjun Ye, Long Chen

机构 * Xiaomi Embodied Intelligence Team（小米具身智能团队） 纠错

专题命中 仿真与规划 ：world model(abstract)；world model(abstract)；分类 cs.CV、cs.RO

AI总结 OneVL通过双重辅助解码器监督的紧凑潜在标记，实现了视觉-语言解释的一步潜在推理与规划，首次在延迟条件下超越了显式推理方法。

Comments Technical Report; 49 pages, 22 figures, 10 tables; Project Page at https://xiaomi-embodied-intelligence.github.io/OneVL GitHub at https://github.com/xiaomi-research/onevl

Data-Asymmetric Latent Imagination and Reranking for 3D Robotic Imitation Learning

数据不对称潜在想象与重排序在3D机器人模仿学习中的应用

Lianghao Luo, Xizhou Bu, Ruyan Liu, Qingqiu Huang, Chufeng Tang, Xiaoshuai Hao, Hongbo Wang, Wei Li

专题命中 仿真与规划 ：world model(abstract)；world model(abstract)；分类 cs.RO

AI总结 本文提出DALI-R框架，通过潜在世界模型和任务完成评分器，利用混合质量轨迹提升3D机器人模仿学习的决策能力，无需额外高质量示范。