arXiv preprint 2026 / 占用预测

SparseOccVLA: Bridging Occupancy and Vision-Language Models via Sparse Queries for Unified 4D Scene Understanding and Planning

作者：Chenxu Dang, Jie Wang, Guang Li, Zhiwen Hou, Zihan You, Hangjun Ye, Jie Ma, Long Chen, Yan Wang · 单位：Huazhong University of Science and Technology, Xiaomi EV Institute for AI Industry Research (AIR), Tsinghua University · 会议/期刊：arXiv preprint 2026 · 方向：占用预测 · 发布日期：2026-01-10

世界模型3D/4D建模占用预测规划控制
新标签打开 AlphaXiv新标签打开 Project新标签打开 Code
SparseOccVLA: Bridging Occupancy and Vision-Language Models via Sparse Queries for Unified 4D Scene Understanding and Planning figure
AlphaXiv 中文概览（可滚动查看）