[Under Review] 🌍 Any 3D Scene is Worth 1K Tokens: 3D-Grounded Representation for Scene Generation at Scale

Dongxu Wei*  ·  Qi Xu*  ·  Zhiqi Li  ·  Hangning Zhou†  ·  Cong Qiu  · 
Hailong Qin  ·  Mu Yang  ·  Zhaopeng Cui  ·  Peidong Liu✉️

Paper | Project Page

If you find this repository useful, please give us a star🌟!

teaser.mp4

News

• 2026/4/13: Our paper is available on arXiv. Code will be released soon. Stay tuned!

TODO

• Release code

Citation

If you find this useful, please consider citing:

@article{wei20263drae,
  author    = {Wei, Dongxu and Xu, Qi and Li, Zhiqi and Zhou, Hangning and Qiu, Cong and Qin, Hailong and Yang, Mu and Cui, Zhaopeng and Liu, Peidong},
  title     = {Any 3D Scene is Worth 1K Tokens: 3D-Grounded Representation for Scene Generation at Scale},
  journal   = {arXiv},
  year      = {2026},
}

Acknowledgments

Our code implementation is greatly inspired by the following outstanding contributions to the open-source community: RAE, LVSM, SEVA, DepthAnything3, VGGT, DINOv2, SigLIP2.