Stream4D: 4D-Consistency for Streaming Autoregressive Diffusion Video Models

Hugging Face Daily Papers Papers

Summary

Stream4D enhances streaming autoregressive diffusion video models by introducing a dynamic 4D reconstruction reward and motion prior to maintain coherent motion and reduce geometric drift.

Streaming autoregressive diffusion models enable real-time, long-horizon video generation, but their training objectives optimize local frame prediction rather than the geometry and dynamics of a coherent world: long rollouts accumulate geometric drift and degrade into static or unnatural motion. Recent bidirectional approaches address this problem using rewards signals built upon 3D Gaussian-Splatting reconstruction. However, a single rigid 3d reconstruction cannot model a dynamic scene, so this critic penalizes genuine object motion as reconstruction error and is maximized by freezing the video. This shortcut is especially detrimental in the AR setting, where each chunk can propagate an already-static configuration. In this work, we propose Stream4D, which replaces the static critic with a feed-forward 4D reconstruction reward that explicitly models scene dynamics, allowing coherent motion to receive high consistency rewards. To further guide motion magnitude and quality, we add a motion prior that rewards natural scene-flow magnitude while penalizing jitter and non-rigid artifacts. Our final recipe combines these two terms with a lightweight perceptual anchor. Across various autoregressive video backbones and various generation horizons, Stream4D improves 4D reconstruction quality, preserves motion more effectively, and achieves higher human-aligned preference. Project page: https://banyuanhao.github.io/Stream4D/
Original Article
View Cached Full Text

Cached at: 08/27/26, 07:19 AM

Paper page - Stream4D: 4D-Consistency for Streaming Autoregressive Diffusion Video Models

Source: https://huggingface.co/papers/2608.19556

Abstract

Stream4D improves autoregressive video generation by replacing static 3D critics with a dynamic 4D reconstruction reward and motion prior to preserve coherent motion and reduce geometric drift.

Streamingautoregressive diffusion modelsenable real-time, long-horizon video generation, but their training objectives optimize local frame prediction rather than the geometry and dynamics of a coherent world: long rollouts accumulate geometric drift and degrade into static or unnatural motion. Recent bidirectional approaches address this problem using rewards signals built upon3D Gaussian-Splattingreconstruction. However, a single rigid 3d reconstruction cannot model a dynamic scene, so this critic penalizes genuine object motion as reconstruction error and is maximized by freezing the video. This shortcut is especially detrimental in the AR setting, where each chunk can propagate an already-static configuration. In this work, we propose Stream4D, which replaces the static critic with afeed-forward 4D reconstruction rewardthat explicitly models scene dynamics, allowing coherent motion to receive high consistency rewards. To further guide motion magnitude and quality, we add amotion priorthat rewards naturalscene-flowmagnitude while penalizing jitter and non-rigid artifacts. Our final recipe combines these two terms with a lightweightperceptual anchor. Across various autoregressive video backbones and various generation horizons, Stream4D improves4D reconstructionquality, preserves motion more effectively, and achieves higher human-aligned preference. Project page: https://banyuanhao.github.io/Stream4D/

View arXiv pageView PDFProject pageAdd to collection

Get this paper in your agent:

hf papers read 2608\.19556

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/2608.19556 in a model README.md to link it from this page.

Datasets citing this paper0

No dataset linking this paper

Cite arxiv.org/abs/2608.19556 in a dataset README.md to link it from this page.

Spaces citing this paper0

No Space linking this paper

Cite arxiv.org/abs/2608.19556 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.