Scalable Non-Equivariant 3D Molecule Generation via Rotational Alignment

Yuhui Ding, Thomas Hofmann
Proceedings of the 42nd International Conference on Machine Learning, PMLR 267:13814-13824, 2025.

Abstract

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

Cite this Paper

BibTeX
@InProceedings{pmlr-v267-ding25a, title = {Scalable Non-Equivariant 3{D} Molecule Generation via Rotational Alignment}, author = {Ding, Yuhui and Hofmann, Thomas}, booktitle = {Proceedings of the 42nd International Conference on Machine Learning}, pages = {13814--13824}, year = {2025}, editor = {Singh, Aarti and Fazel, Maryam and Hsu, Daniel and Lacoste-Julien, Simon and Berkenkamp, Felix and Maharaj, Tegan and Wagstaff, Kiri and Zhu, Jerry}, volume = {267}, series = {Proceedings of Machine Learning Research}, month = {13--19 Jul}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v267/main/assets/ding25a/ding25a.pdf}, url = {https://proceedings.mlr.press/v267/ding25a.html}, abstract = {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} }
Endnote
%0 Conference Paper %T Scalable Non-Equivariant 3D Molecule Generation via Rotational Alignment %A Yuhui Ding %A Thomas Hofmann %B Proceedings of the 42nd International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2025 %E Aarti Singh %E Maryam Fazel %E Daniel Hsu %E Simon Lacoste-Julien %E Felix Berkenkamp %E Tegan Maharaj %E Kiri Wagstaff %E Jerry Zhu %F pmlr-v267-ding25a %I PMLR %P 13814--13824 %U https://proceedings.mlr.press/v267/ding25a.html %V 267 %X 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
APA
Ding, Y. & Hofmann, T.. (2025). Scalable Non-Equivariant 3D Molecule Generation via Rotational Alignment. Proceedings of the 42nd International Conference on Machine Learning, in Proceedings of Machine Learning Research 267:13814-13824 Available from https://proceedings.mlr.press/v267/ding25a.html.

Related Material