Scaffold Diffusion: Sparse Multi-Category Voxel Structure Generation with Discrete Diffusion
Fuente:
arXiv
Saved in:
| Main Author: | |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866914167952244736 |
|---|---|
| author | Jung, Justin |
| author_facet | Jung, Justin |
| contents | Generating realistic sparse multi-category 3D voxel structures is difficult due to the cubic memory scaling of voxel structures and moreover the significant class imbalance caused by sparsity. We introduce Scaffold Diffusion, a generative model designed for sparse multi-category 3D voxel structures. By treating voxels as tokens, Scaffold Diffusion uses a discrete diffusion language model to generate 3D voxel structures. We show that discrete diffusion language models can be extended beyond inherently sequential domains such as text to generate spatially coherent 3D structures. We evaluate on Minecraft house structures from the 3D-Craft dataset and demonstrate that, unlike prior baselines and an auto-regressive formulation, Scaffold Diffusion produces realistic and coherent structures even when trained on data with over 98% sparsity. We provide an interactive viewer where readers can visualize generated samples and the generation process: https://scaffold.deepexploration.org/ |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2509_00062 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Scaffold Diffusion: Sparse Multi-Category Voxel Structure Generation with Discrete Diffusion Jung, Justin Computer Vision and Pattern Recognition Artificial Intelligence Machine Learning Generating realistic sparse multi-category 3D voxel structures is difficult due to the cubic memory scaling of voxel structures and moreover the significant class imbalance caused by sparsity. We introduce Scaffold Diffusion, a generative model designed for sparse multi-category 3D voxel structures. By treating voxels as tokens, Scaffold Diffusion uses a discrete diffusion language model to generate 3D voxel structures. We show that discrete diffusion language models can be extended beyond inherently sequential domains such as text to generate spatially coherent 3D structures. We evaluate on Minecraft house structures from the 3D-Craft dataset and demonstrate that, unlike prior baselines and an auto-regressive formulation, Scaffold Diffusion produces realistic and coherent structures even when trained on data with over 98% sparsity. We provide an interactive viewer where readers can visualize generated samples and the generation process: https://scaffold.deepexploration.org/ |
| title | Scaffold Diffusion: Sparse Multi-Category Voxel Structure Generation with Discrete Diffusion |
| topic | Computer Vision and Pattern Recognition Artificial Intelligence Machine Learning |
| url | https://arxiv.org/abs/2509.00062 |