One-Forcing: Towards Stable One-Step Autoregressive Video Generation

Hugging Face Daily Papers Papers
autoregressive-video-generation one-step-generation gan-loss consistency-distillation video-generation

Summary

One-Forcing improves one-step video generation by augmenting the DMD objective with an auxiliary GAN loss, achieving state-of-the-art performance with reduced training costs.

Recent advances have substantially improved real-time interactive video generation in the autoregressive regime. However, most existing few-step autoregressive video generation methods, often distilled from a corresponding many-step teacher, default to a 4-step sampling configuration, which still incurs considerable latency during deployment and suffers from severe quality degradation when the number of sampling steps is further reduced, particularly in the one-step setting. Trajectory-style consistency distillation methods often produce videos with weak dynamics, while DMD-based approaches, such as Self-Forcing, tend to yield blurry frames. To address this challenge, we propose One-Forcing, a simple yet effective approach which augments the DMD objective with an auxiliary GAN loss for high-quality and efficient one-step video generation. Experiments on VBench show that One-Forcing achieves a total score of 83.76, establishing state-of-the-art performance among one-step causal video generation methods and remaining competitive with strong many-step approaches. We further demonstrate that one-step framewise autoregressive generation can be achieved stably with merely one-third of the training cost of the chunkwise model, a setting that prior methods have failed to achieve successfully.
Original Article
View Cached Full Text

Cached at: 06/01/26, 07:21 PM

Paper page - One-Forcing: Towards Stable One-Step Autoregressive Video Generation

Source: https://huggingface.co/papers/2605.23458

Abstract

One-Forcing improves one-step video generation quality and efficiency by combining DMD objective with GAN loss, achieving state-of-the-art results with reduced training costs.

Recent advances have substantially improved real-time interactive video generation in the autoregressive regime. However, most existing few-stepautoregressive video generationmethods, often distilled from a corresponding many-step teacher, default to a 4-step sampling configuration, which still incurs considerable latency during deployment and suffers from severe quality degradation when the number of sampling steps is further reduced, particularly in the one-step setting.Trajectory-style consistency distillationmethods often produce videos with weak dynamics, whileDMD-based approaches, such asSelf-Forcing, tend to yield blurry frames. To address this challenge, we propose One-Forcing, a simple yet effective approach which augments the DMD objective with an auxiliaryGAN lossfor high-quality and efficientone-step video generation. Experiments on VBench show that One-Forcing achieves a total score of 83.76, establishing state-of-the-art performance among one-stepcausal video generationmethods and remaining competitive with strong many-step approaches. We further demonstrate that one-step framewise autoregressive generation can be achieved stably with merely one-third of the training cost of thechunkwise model, a setting that prior methods have failed to achieve successfully.

View arXiv pageView PDFProject pageGitHub30Add to collection

Get this paper in your agent:

hf papers read 2605\.23458

Don’t have the latest CLI?curl \-LsSf https://hf\.co/cli/install\.sh \| bash

Models citing this paper0

No model linking this paper

Cite arxiv.org/abs/2605.23458 in a model README.md to link it from this page.

Datasets citing this paper0

No dataset linking this paper

Cite arxiv.org/abs/2605.23458 in a dataset README.md to link it from this page.

Spaces citing this paper0

No Space linking this paper

Cite arxiv.org/abs/2605.23458 in a Space README.md to link it from this page.

Collections including this paper0

No Collection including this paper

Add this paper to acollectionto link it from this page.

Similar Articles

Streaming Video Generation with Streaming Force Control

Hugging Face Daily Papers

StreamForce is a causal, unified video generation model that provides real-time, physically grounded responses to time-varying forces through a distillation pipeline and autoregressive architecture, achieving state-of-the-art performance in force adherence and motion realism.