跳到正文
HuggingFace Daily Papers(社區熱門論文)·· 2026-10-15AI 評分43

Velox:空間記憶提升世界模型效率

Velox: Spatial Memory Makes World Models Efficient

AI 導讀

Velox 提出免訓練注意力路由器,利用幾何視錐切片選取區域上下文,加速長影片生成並維持視覺連貫性。在 chunk 48 時,注意力加速 4.36 倍,去噪加速 2.43 倍。在完整 158-case WBench navigation split 上,其 API-free 平均分與 dense LingBot-Fast-Cam 相差不超過 0.15 分。

正文

Published on Oct 15

Authors:

,

Abstract

Velox accelerates long video rollouts by routing attention via geometric frustum slabs, reducing latency while preserving visual coherence.

Video world models turn actions and camera motion into long visual rollouts, making video a natural interface for interactive simulation. Such interaction requires responsive generation, yet coherent rollouts must retain visual memory of what has already been generated. Chunked video diffusion models usually keep this context in growing KV caches, so later chunks become progressively slower. To address this bottleneck, we introduce Velox, a training-free attention router for region-level context selection using frustum slab correspondences from streaming geometry. Velox represents historical blocks as camera frustum slabs with depth intervals and obtains current-block depth through recency-priority multiwarp. For well-covered regions, frustum slab overlap softly modulates each layer's model-derived block importance. Uncovered regions retain the model-derived route without geometric bias. The fixed token-block granularity matches FlashAttention-style tiled execution and reduces the attention term that grows with rollout length. Frustum slab construction runs asynchronously with denoising on the same H200. At chunk 48, Velox accelerates attention by 4.36times and denoising by 2.43times. On the full 158-case WBench navigation split, it remains within 0.15 API-free average points of dense LingBot-Fast-Cam. Please see our project page at https://velox-world.com/velox/ for more details.

View arXiv page View PDF Project page Add to collection

Get this paper in your agent:

hf papers read 2610.22311

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/2610.22311 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/2610.22311 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/2610.22311 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.

來源:HuggingFace Daily Papers(社區熱門論文) · huggingface.co