Glance Forcing:讓雙向擴散模型實現自回歸生成
Glance Forcing: Making Bidirectional Diffusion Autoregressive at a Glance
研究提出 Glance Forcing,將雙向擴散模型蒸餾為自回歸生成模型,僅用一條 PF-ODE 軌跡便超越已充分訓練的 SOTA 基線。VBench Total 由 84.04 升至 84.22,軌跡收集成本由數千 GPU-hours 降至約一個 A100-GPU-hour。方法將去噪軌跡拆分為慢、快兩個專家,並保持預訓練教師模型完全凍結。
Published on Oct 14
Authors:
,
,
,
,
Abstract
Autoregressive generation underlies real-time, causal video synthesis and is the backbone of video world models. Today's dominant approach distills a bidirectional diffusion model into a causal few-step student, but existing pipelines make this conversion prohibitively expensive, requiring offline Probability Flow ODE} (PF-ODE) trajectories that cost tens of thousands of GPU-hours and terabytes of storage. We show it can be made dramatically more efficient. Pushing this efficiency to its extreme, we reduce trajectory supervision to a single sample, the strictest setting possible, and find that it causes severe quality collapse. We trace this to the teacher's trajectory geometry: the high-noise regime, governing semantic layout and motion, and the low-noise regime, governing fine texture exhibit fundamentally different curvature and optimization behavior, so a single monolithic student cannot jointly fit both, causing severe representation interference. Motivated by this, we propose Glance Forcing, a lightweight distillation framework that decouples the denoising trajectory into two complementary slow and fast experts, each guided by a tailored, stage-specific objective, while keeping the pretrained teacher entirely frozen. This decoupling further reveals a general mechanism underlying few-step distillation: PF-ODE distillation mainly straightens the few-step trajectory geometry, while the subsequent distribution-matching stage independently governs distributional quality. Using only one PF-ODE trajectory, Glance Forcing already surpasses the fully-trained state-of-the-art baseline, raising the VBench Total score from 84.04 to 84.22 while cutting trajectory-collection cost from thousands of GPU-hours to about one A100-GPU-hour, with markedly larger gains in dynamic degree, VisionReward, and instruction-following, and consistent preference from human raters over strong baselines.
View arXiv page View PDF Project page GitHub 3 Add to collection
Get this paper in your agent:
hf papers read 2610.20220
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/2610.20220 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/2610.20220 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/2610.20220 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