Code for AssembleFlow, ICLR 2025
-
Updated
Mar 11, 2025 - Python
Code for AssembleFlow, ICLR 2025
SE(3) and E(n) equivariant graph neural network message passing kernel preserving exact 3D rotational and translational symmetries in molecular and physical systems.
SE(3) and E(n) equivariant graph neural network message passing kernel preserving exact 3D rotational and translational symmetries in molecular and physical systems.
Contact-grounded SE(3)-equivariant dexterous grasp generative flows
E(3)-equivariant GNNs for protein-ligand binding-affinity prediction (EGNN + e3nn tensor-product), with a machine-precision invariance test suite.
To associate your repository with the se3-equivariance topic, visit your repo's landing page and select "manage topics."