Sicheng Zuo, Zixun Xie, Wenzhao Zheng, Shaoqing Xu, Fang Li, Hanbing Li, Long Chen, Zhi-Xin Yang, Jiwen Lu

Researchers propose DVGT-2, a streaming vision-geometry-action model for autonomous driving that processes camera inputs online rather than in batches. Unlike recent vision-language-action approaches that use language as an auxiliary task, DVGT-2 prioritizes dense 3D geometry reconstruction as the primary signal for planning decisions. The model uses temporal causal attention and historical feature caching to enable real-time inference while achieving better geometry reconstruction than prior methods, and generalizes across different camera configurations without fine-tuning on both closed-loop and open-loop benchmarks.