cs.LGMay 29, 2025

Equivariant Spherical Transformer for Efficient Molecular Modeling

Authors: Junyi An, Xinyu Lu, Yunfei Shi, Peijia Lin, Licheng Xu, Qianwei Tang, Chao Qu, Fenglei Cao, 

Organizations: SAIS · Shanghai Innovation Institute · Xiamen University · Sun Yat-sen University · 

Abstract

Equivariant Graph Neural Networks (GNNs) have significantly advanced the modeling of 3D molecular structure by leveraging group representations. However, message passing built only from exactly equivariant operations can have restricted practical expressiveness at a fixed representation degree. We introduce the Equivariant Spherical Transformer (EST), a plug-and-play module that applies Transformer attention to spherical spatial representations obtained from group representations. EST explicitly mixes information across sampled orientations, unlike prior orientation-wise spherical operations, while retaining a strong approximate equivariant inductive bias through near-uniform spherical sampling. As demonstrated by our experiments on challenging benchmarks like OC20, MPtrj and QM9, EST-based models achieve state-of-the-art performance. For the complex molecular systems within OC20, small models empowered by EST can outperform some larger models and those using additional data. We characterize the ideal closed-domain condition under which the continuous operation is exactly equivariant and empirically quantify the finite-sampling accuracy--equivariance--memory trade-off used in practice.

Figures & tables

Appendix figures & tables21 assets

Supplementary material from the paper’s appendix.

Appendix

Explore similar work

CardsList
  1. EquiformerV3: Scaling Efficient, Expressive, and General SE(3)-Equivariant Graph Attention Transformers

    Apr 10, 2026Yi-Lun Liao, Alexander J. Hoffman, Sabrina C. Shen +3Equivariant GNNsGraph Transformer

  2. NEAT: Neighborhood-Guided, Efficient, Autoregressive Set Transformer for 3D Molecular Generation

    Dec 5, 2025Daniel Rose, Roxane Axel Jacob, Johannes Kirchmair +13D Molecular GenerationAutoregressive Transformer

  3. GEqTrain: A Configuration-Driven Framework for Retargeting Equivariant Graph Neural Networks Across 3D Scientific Tasks

    Jul 21, 2026Daniele Angioletti, Marco Nobile, Vittorio LimongelliFlow MatchingGenerative Modeling