通过旋转对齐实现高效3D分子生成，性能媲美顶尖模型
Scalable Non-Equivariant 3D Molecule Generation via Rotational Alignment
- 为每个分子学习独立的SO(3)旋转，构建对齐潜在空间
- 生成质量接近顶尖等变模型，训练与采样效率更高
- 适合追求高效分子生成的研究者和工业应用
等变扩散模型在3D分子生成中表现优异，其通过SE(3)等变去噪网络引入三维分子的欧几里得对称性。然而，专用的等变架构限制了扩散模型的可扩展性和效率。本文提出一种放松等变约束的方法：为每个分子学习样本相关的SO(3)变换，构建对齐的潜在空间，并在此基础上训练非等变扩散模型。实验表明，该方法显著优于此前的非等变模型，在生成质量上达到顶尖等变模型水平，同时具备更优的训练与采样效率。代码已开源：https://github.com/skeletondyh/RADM。
原文摘要 · Abstract (English)
Equivariant diffusion models have achieved impressive performance in 3D molecule generation. These models incorporate Euclidean symmetries of 3D molecules by utilizing an SE(3)-equivariant denoising network. However, specialized equivariant architectures limit the scalability and efficiency of diffusion models. In this paper, we propose an approach that relaxes such equivariance constraints. Specifically, our approach learns a sample-dependent SO(3) transformation for each molecule to construct an aligned latent space. A non-equivariant diffusion model is then trained over the aligned representations. Experimental results demonstrate that our approach performs significantly better than previously reported non-equivariant models. It yields sample quality comparable to state-of-the-art equivariant diffusion models and offers improved training and sampling efficiency. Our code is available at https://github.com/skeletondyh/RADM
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。