VFlash：通过 Flash 基元解码实现快速、精确的视觉规划
VFlash: Fast, Accurate Visual Planning via Flash Primitive Decoding
VFlash 通过 Flash 基元解码实现快速、精确的视觉规划，其轻量级绘图器在单次前向传递中生成整个基元，在 3D 点运动预测任务中速度提升约 25 倍，并在 MolmoSpaces 四个任务套件上达到 70.2% 的平均成功率。
Published on Nov 5
Authors:
,
,
,
,
,
,
,
,
,
,
,
,
,
,
,
,
,
Abstract
Models for physical AI tasks, such as robotics and autonomous driving, need not reason solely in language. They can use visual plans--points, bounding boxes, and trajectories--to represent spatial goals and anticipate motion. However, autoregressive vision-language models generate these plans one coordinate token at a time, making richer predictions increasingly costly. We introduce VFlash, a method for fast, accurate visual planning through Flash Primitive Decoding. VFlash exploits the known structure of visual primitives to predict their coordinates jointly, using a lightweight drafter that generates an entire primitive in one forward pass. The layout adapts to the primitive's type, dimensionality, and length, supporting compact boxes and long, variable-length trajectories. Crucially, spatial accuracy need not require reproducing an autoregressive model's exact token sequence. VFlash therefore trains on annotated primitives and directly accepts structurally valid predictions without target-token agreement. Across the evaluated 2D and 3D robot trajectory tasks, VFlash consistently reduces prediction errors relative to autoregressive decoding. Preliminary single-rollout measurements indicate approximately 25x faster generation for 3D point motion forecasting. A world action model conditioned on VFlash plans achieves 70.2% average success across four MolmoSpaces task suites, compared with 62.6% for the baseline.
View arXiv page View PDF Project page Add to collection
Get this paper in your agent:
hf papers read 2611.08449
Don't have the latest CLI?
curl -LsSf https://hf.co/cli/install.sh | bash
Models citing this paper 0
No model linking this paper
Cite arxiv.org/abs/2611.08449 in a model README.md to link it from this page.
Datasets citing this paper 0
No dataset linking this paper
Cite arxiv.org/abs/2611.08449 in a dataset README.md to link it from this page.
Spaces citing this paper 0
No Space linking this paper
Cite arxiv.org/abs/2611.08449 in a Space README.md to link it from this page.
Collections including this paper 0
No Collection including this paper
Add this paper to a collection to link it from this page.
来源: Hugging Face Daily Papers · huggingface.co