Straight-Line Diffusion Model for Efficient 3D Molecular Generation

Yuyan Ni, Shikun Feng, Haohan Chi, Bowen Zheng, Huan-ang Gao +3 more
3/4/2025
cs.LGcs.AI

No official code found. Best implementation: fengshikun/SLDM (2 stars, MIT, last commit 2025-04-02).

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.

View on arXivView PDF

Code Implementations(1)

fengshikun/SLDMCommunity100%
MIT
21Python, Jupyter NotebookMar 22, 20251 years agoMIT

Similar Papers

LMDM:Latent Molecular Diffusion Model For 3D Molecule Generation

86% match2024

Geometric-Facilitated Denoising Diffusion Model for 3D Molecule Generation

85% match1 repo102024

Equivariant Neural Diffusion for Molecule Generation

85% match1 repo92026

Bridging the Gap between Learning and Inference for Diffusion-Based Molecule Generation

85% match1 repo2024

Geometric Latent Diffusion Models for 3D Molecule Generation

85% match1 repo2762023

Breaking the Bottlenecks: Scalable Diffusion Models for 3D Molecular Generation

85% match2026

Cite this paper

@article{ni2025straightline,
  title  = {Straight-Line Diffusion Model for Efficient 3D Molecular Generation},
  author = {Yuyan Ni and Shikun Feng and Haohan Chi and Bowen Zheng and Huan-ang Gao and Wei-Ying Ma and Zhi-Ming Ma and Yanyan Lan},
  year   = {2025},
  eprint = {2503.02918},
  archivePrefix = {arXiv},
  url    = {http://arxiv.org/abs/2503.02918v2}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/2503.02918.svg)](https://codeofpaper.com/paper/2503.02918)

Discussion