Paper page - DreamX-World 1.0: A General-Purpose Interactive World Model

Source: https://huggingface.co/papers/2606.16993 Authors:

,

,

,

,

,

,

,

,

,

,

,

,

,

,

,

,

,

,

,

,

Abstract

DreamX-World 1.0 is a interactive text/image-to-video model that generates long-horizon content with camera control and scene persistence using specialized encoding, training techniques, and optimization methods.

DreamX-World 1.0 is a general-purpose interactive text/image-to-video world model for controllable long-horizon generation. It supports camera navigation, revisits to previously observed regions, and promptable events across photorealistic, game-style, and stylized domains. Our data engine combines camera-accurate Unreal Engine rendering, action-rich gameplay recordings, and real-world videos with recovered camera geometry. For camera control, we introduceE-PRoPE, a lightweight variant ofprojective positional encodingthat retainsPRoPE’s projective camera geometry while applyingcamera-aware attentionto spatially reduced tokens. We convert abidirectional video generatorinto a few-stepautoregressive world modelusingcausal forcing,DMD-style distillation, andlong-rollout training. Training on self-generated long-horizon contexts exposes the model to its own generated history and reduces the style and color drift that accumulates across autoregressive chunks.Memory-Conditioned Scene Persistenceretrieves earlier views through camera-geometry-based retrieval, whileresidual recyclingmakes the conditioning path less sensitive to imperfect memory latents.Event Instruction Tuningadds composable event control, andreinforcement learning alignmentrecovers camera control and visual quality after distillation. Withmixed-precision DiT execution, residual reuse, 75\%-pruned VAE decoding, andasynchronous pipeline parallelism, DreamX-World 1.0 reaches up to 16\,FPS on eight RTX\,5090 GPUs. On our 5-second basic evaluation, DreamX-World 1.0 achieves a camera-control score of 73.75 and an overall score of 84.76, outperforming HY-WorldPlay 1.5 and LingBot-World in overall score, which achieve 80.79 and 80.45, respectively.

View arXiv pageView PDFProject pageGitHub264Add to collection

Get this paper in your agent:

hf papers read 2606\.16993

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

Models citing this paper1

#### GD-ML/DreamX-World-5B Image-to-Video• 5B• Updatedabout 4 hours ago • 1

Datasets citing this paper0

No dataset linking this paper

Cite arxiv.org/abs/2606.16993 in a dataset README.md to link it from this page.

Spaces citing this paper0

No Space linking this paper

Cite arxiv.org/abs/2606.16993 in a Space README.md to link it from this page.

Collections including this paper2