Compositional Visual Planning via Inference-Time Diffusion Scaling

[Project page] [Paper] [ArXiv]

Yixin Zhang¹, Yunhao Luo^{1, 2}, Utkarsh Mishra¹, Woo Chul Shin¹, Yongxin Chen¹, Danfei Xu¹,

¹Georgia Tech, ²University of Michigan

This is the official implementation for "Compositional Visual Planning via Inference-Time Diffusion Scaling".

🚀 Code Release TODOs

• ✅ Project Website Release
• ⏳ Algorithm Code Release
• ⏳ Pre-trained Model Release
• ⏳ Environment Code Release

🙏 Acknowledgement

• The implementation of DDPM is adapted from CleanDiffuser.

Contact Yixin Zhang if you have any questions or suggestions.

📝 Citations

If you find our work useful, please consider citing:

@misc{zhang2026compositionalvisualplanninginferencetime,
      title={Compositional Visual Planning via Inference-Time Diffusion Scaling}, 
      author={Yixin Zhang and Yunhao Luo and Utkarsh Aashu Mishra and Woo Chul Shin and Yongxin Chen and Danfei Xu},
      year={2026},
      eprint={2603.02646},
      archivePrefix={arXiv},
      primaryClass={cs.RO},
      url={https://arxiv.org/abs/2603.02646}, 
}