Straight-Line Diffusion Model for Efficient 3D Molecular Generation
Abstract
Diffusion-based models have shown great promise in molecular generation but often require a large number of sampling steps to generate valid samples. In this paper, we introduce a novel Straight-Line Diffusion Model (SLDM) to tackle this problem, by formulating the diffusion process to follow a linear trajectory. The proposed process aligns well with the noise sensitivity characteristic of molecular structures and uniformly distributes reconstruction effort across the generative process, thus enhancing learning efficiency and efficacy. Consequently, SLDM achieves state-of-the-art performance on 3D molecule generation benchmarks, delivering a 100-fold improvement in sampling efficiency.
Cite
Text
Ni et al. "Straight-Line Diffusion Model for Efficient 3D Molecular Generation." Advances in Neural Information Processing Systems, 2025.Markdown
[Ni et al. "Straight-Line Diffusion Model for Efficient 3D Molecular Generation." Advances in Neural Information Processing Systems, 2025.](https://mlanthology.org/neurips/2025/ni2025neurips-straightline/)BibTeX
@inproceedings{ni2025neurips-straightline,
title = {{Straight-Line Diffusion Model for Efficient 3D Molecular Generation}},
author = {Ni, Yuyan and Feng, Shikun and Chi, Haohan and Zheng, Bowen and Gao, Huan-ang and Ma, Wei-Ying and Ma, Zhi-Ming and Lan, Yanyan},
booktitle = {Advances in Neural Information Processing Systems},
year = {2025},
url = {https://mlanthology.org/neurips/2025/ni2025neurips-straightline/}
}