A plain causal transformer that tokenizes atom positions in local frames generates 3D molecules directly; RL against an xTB relaxation reward lifts topology-preserving valid yield from ~50% to ~95%.
Deep generative models for 3D linker design.Journal of Chemical Information and Modeling, 60(4):1983–1995, 2020
1 Pith paper cite this work, alongside 268 external citations. Polarity classification is still indexing.
1
Pith paper citing it
268
external citations · OpenAlex
fields
physics.comp-ph 1years
2026 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Atomic Design Transformer: Scaffold-Conditioned 3D Molecule Generation via xTB-Reward Reinforcement Learning
A plain causal transformer that tokenizes atom positions in local frames generates 3D molecules directly; RL against an xTB relaxation reward lifts topology-preserving valid yield from ~50% to ~95%.