跳到正文
HuggingFace Daily Papers(社區熱門論文)·· 3 天前AI 評分42

SpecWM:將驗證轉化為串流世界模型的精煉

SpecWM: Turning Verification into Refinement for Streaming World Models

AI 導讀

SpecWM 提出免訓練推理框架,利用目標模型的潛在速度場驗證草稿片段,並按門檻決定接受、精煉或拒絕。在 WBench Navigation 上,SpecWM 以 1.63 倍加速提升視覺質素,並將 NavScore 由草稿的 75.99 提高至 85.63。

正文

Published on Nov 19

Authors:

,

,

,

,

,

,

,

,

,

,

Abstract

SpecWM is a training-free inference framework for streaming video world models that verifies draft chunks using the target model's latent velocity field and applies reusable corrections via threshold-based accept/refine/reject decisions to improve visual quality, control, and speed.

Recently, video world models have advanced interactive generation, with large models offering higher quality and smaller models enabling faster inference. Existing speculative methods combine their strengths through fast draft generation and selective target regeneration guided by rewards on decoded frames. However, verification covers only the external scorer's evaluation dimensions and adds decoding and scoring costs. Moreover, these scores provide no correction, requiring costly target regeneration for rejected drafts. To address these limitations, we propose Speculative World Model (SpecWM), a unified, training-free inference framework for streaming world models. It verifies draft chunks using the target's latent velocity field. Verification also yields a reusable correction; two thresholds select Accept, Refine, or Reject. Specifically, one-step refinement reuses it to improve perceptual quality, while three-step refinement strengthens camera following and trajectory consistency. These findings motivate transition-aware refinement. On WBench Navigation, SpecWM improves visual quality at 1.63times speedup over the target and raises NavScore from the draft's 75.99 to 85.63, achieving the best quality-speed-control trade-off. Additionally, our LongSpecBench shows sustained quality and control over long, multi-turn rollouts, while experiments on Hy-Nomas validate SpecWM across backbones and denoising budgets.

View arXiv page View PDF Add to collection

Get this paper in your agent:

hf papers read 2611.26211

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/2611.26211 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/2611.26211 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/2611.26211 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