AtomComposer uses online RL with multi-composition training to discover up to 10x more valid 3D isomers on unseen chemical formulas than single-composition baselines.
hub
Enum eration of 166 billion organic small molecules in the chemical universe database GDB -17
3 Pith papers cite this work, alongside 1,550 external citations. Polarity classification is still indexing.
3
Pith papers citing it
1,550
external citations · OpenAlex
hub tools
fields
cs.LG 3verdicts
UNVERDICTED 3representative citing papers
Multigrid training accelerates convergence and improves generalization for receptor-conditioned 3D ligand generation by transferring parameters from coarse to fine graph and voxel resolutions.
Bayesian E(3)-equivariant MLPs with joint energy-force NLL loss achieve competitive accuracy while enabling uncertainty-guided active learning, OOD detection, and calibration.
citing papers explorer
-
AtomComposer: Discovering Chemical Space from First Principles with Reinforcement Learning
AtomComposer uses online RL with multi-composition training to discover up to 10x more valid 3D isomers on unseen chemical formulas than single-composition baselines.
-
Multigrid Training for Molecular Generation using Graph Neural Networks
Multigrid training accelerates convergence and improves generalization for receptor-conditioned 3D ligand generation by transferring parameters from coarse to fine graph and voxel resolutions.
-
Bayesian E(3)-Equivariant Interatomic Potential with Iterative Restratification of Many-body Message Passing
Bayesian E(3)-equivariant MLPs with joint energy-force NLL loss achieve competitive accuracy while enabling uncertainty-guided active learning, OOD detection, and calibration.