跳到正文
HuggingFace Daily Papers(社區熱門論文)·· 2026-10-08AI 評分37

ForcingBooster:快速自回歸影片擴散的自適應逐區塊去噪

ForcingBooster: Adaptive Patch-wise Denoising for Fast Autoregressive Video Diffusion

AI 導讀

ForcingBooster 按影片區塊的難度動態分配去噪步數,將區塊上的總去噪步數減少約 50%,並維持與 4 步基線相若的視覺質素。它兼容大部分現有的少步自回歸擴散模型,只需輕量級路由模組及適度訓練。

正文

Published on Oct 8

Authors:

,

,

,

,

Abstract

A framework dynamically assigns denoising steps to video patches based on difficulty, cutting computation by half while preserving quality.

Autoregressive video diffusion models have become a leading approach to real-time interactive video generation and video world modeling. However, existing methods typically allocate computation uniformly across space when generating the next video chunk, updating all patches with the same denoising schedule and number of function evaluations. This ignores the spatial heterogeneity of video content: static, simple, or recurring regions are often easier to denoise, while other regions require further refinement or additional context. A uniform budget thus leads to redundant computation on easy patches, whereas reducing the fixed step count can leave difficult patches insufficiently refined. We introduce , a general framework that dynamically allocates denoising steps to individual patches: easy patches stop early to improve efficiency, while difficult patches receive more refinement steps. This design accelerates generation with negligible quality degradation. Our method is compatible with most existing few-step autoregressive diffusion models, requiring only a lightweight routing module with modest training. Experiments demonstrate that reduces the total number of denoising steps applied across video patches by about 50\% while maintaining visual quality comparable to the four-step baseline.

View arXiv page View PDF Project page GitHub Add to collection

Get this paper in your agent:

hf papers read 2610.12625

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/2610.12625 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/2610.12625 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/2610.12625 in a Space README.md to link it from this page.

Collections including this paper 1

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