NVIDIA 论文提出 VFlash：并行生成视觉规划，机器人规划延迟降低约 80%
摘要
NVIDIA 的新论文 VFlash 提出用轻量扩散模型并行预测整条轨迹、边界框和 3D 运动，而不是逐个坐标生成视觉规划，使机器人规划延迟降低约 80%，同时提升空间精度。在 MolmoSpaces 上，动作模型使用 VFlash 生成的规划后，机器人任务成功率从 62.6% 提升到 70.2%。论文链接：https://www.alphaxiv.org/abs/2611.08449
摘要由模型根据原文撰写，以原文为准。
正文 · 原文
"VFlash: Fast, Accurate Visual Planning via Flash Primitive Decoding"
Generating visual plans for robotics and autonomous driving models is done one coordinate at a time, making detailed trajectories slow to produce.
NVIDIA's new paper VFlash proposes a lightweight diffusion model to predict entire trajectories, bounding boxes, and 3D motion in parallel, rather than generating each coordinate sequentially.
This reduces robot planning latency by around 80% while improving spatial accuracy.
On MolmoSpaces, robot task success improves from 62.6% to 70.2% when an action model uses VFlash's plans.
来源：alphaXiv · x.com
引用原文
askalphaxiv. "VFlash: Fast, Accurate Visual Planning via Flash Primitive Decoding"[EB/OL]. (2026-11-08)[访问日期]. https://x.com/askalphaxiv/status/2119348608122449939.