SemlaFlow – Efficient 3D Molecular Generation with Latent Attention and Equivariant Flow Matching

Ross Irwin, Alessandro Tibo, Jon Paul Janet, Simon Olsson
Proceedings of The 28th International Conference on Artificial Intelligence and Statistics, PMLR 258:3772-3780, 2025.

Abstract

Methods for jointly generating molecular graphs along with their 3D conformations have gained prominence recently due to their potential impact on structure-based drug design. Current approaches, however, often suffer from very slow sampling times or generate molecules with poor chemical validity. Addressing these limitations, we propose Semla, a scalable E(3)-equivariant message passing architecture. We further introduce an unconditional 3D molecular generation model, SemlaFlow, which is trained using equivariant flow matching to generate a joint distribution over atom types, coordinates, bond types and formal charges. Our model produces state-of-the-art results on benchmark datasets with as few as 20 sampling steps, corresponding to a two order-of-magnitude speedup compared to state-of-the-art. Furthermore, we highlight limitations of current evaluation methods for 3D generation and propose new benchmark metrics for unconditional molecular generators. Finally, using these new metrics, we compare our model’s ability to generate high quality samples against current approaches and further demonstrate SemlaFlow’s strong performance.

Cite this Paper

BibTeX
@InProceedings{pmlr-v258-irwin25a, title = {SemlaFlow – Efficient 3D Molecular Generation with Latent Attention and Equivariant Flow Matching}, author = {Irwin, Ross and Tibo, Alessandro and Janet, Jon Paul and Olsson, Simon}, booktitle = {Proceedings of The 28th International Conference on Artificial Intelligence and Statistics}, pages = {3772--3780}, year = {2025}, editor = {Li, Yingzhen and Mandt, Stephan and Agrawal, Shipra and Khan, Emtiyaz}, volume = {258}, series = {Proceedings of Machine Learning Research}, month = {03--05 May}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v258/main/assets/irwin25a/irwin25a.pdf}, url = {https://proceedings.mlr.press/v258/irwin25a.html}, abstract = {Methods for jointly generating molecular graphs along with their 3D conformations have gained prominence recently due to their potential impact on structure-based drug design. Current approaches, however, often suffer from very slow sampling times or generate molecules with poor chemical validity. Addressing these limitations, we propose Semla, a scalable E(3)-equivariant message passing architecture. We further introduce an unconditional 3D molecular generation model, SemlaFlow, which is trained using equivariant flow matching to generate a joint distribution over atom types, coordinates, bond types and formal charges. Our model produces state-of-the-art results on benchmark datasets with as few as 20 sampling steps, corresponding to a two order-of-magnitude speedup compared to state-of-the-art. Furthermore, we highlight limitations of current evaluation methods for 3D generation and propose new benchmark metrics for unconditional molecular generators. Finally, using these new metrics, we compare our model’s ability to generate high quality samples against current approaches and further demonstrate SemlaFlow’s strong performance.} }
Endnote
%0 Conference Paper %T SemlaFlow – Efficient 3D Molecular Generation with Latent Attention and Equivariant Flow Matching %A Ross Irwin %A Alessandro Tibo %A Jon Paul Janet %A Simon Olsson %B Proceedings of The 28th International Conference on Artificial Intelligence and Statistics %C Proceedings of Machine Learning Research %D 2025 %E Yingzhen Li %E Stephan Mandt %E Shipra Agrawal %E Emtiyaz Khan %F pmlr-v258-irwin25a %I PMLR %P 3772--3780 %U https://proceedings.mlr.press/v258/irwin25a.html %V 258 %X Methods for jointly generating molecular graphs along with their 3D conformations have gained prominence recently due to their potential impact on structure-based drug design. Current approaches, however, often suffer from very slow sampling times or generate molecules with poor chemical validity. Addressing these limitations, we propose Semla, a scalable E(3)-equivariant message passing architecture. We further introduce an unconditional 3D molecular generation model, SemlaFlow, which is trained using equivariant flow matching to generate a joint distribution over atom types, coordinates, bond types and formal charges. Our model produces state-of-the-art results on benchmark datasets with as few as 20 sampling steps, corresponding to a two order-of-magnitude speedup compared to state-of-the-art. Furthermore, we highlight limitations of current evaluation methods for 3D generation and propose new benchmark metrics for unconditional molecular generators. Finally, using these new metrics, we compare our model’s ability to generate high quality samples against current approaches and further demonstrate SemlaFlow’s strong performance.
APA
Irwin, R., Tibo, A., Janet, J.P. & Olsson, S.. (2025). SemlaFlow – Efficient 3D Molecular Generation with Latent Attention and Equivariant Flow Matching. Proceedings of The 28th International Conference on Artificial Intelligence and Statistics, in Proceedings of Machine Learning Research 258:3772-3780 Available from https://proceedings.mlr.press/v258/irwin25a.html.

Related Material