Mol-R1 combines regulated in-context distillation (PRID) with iterative supervised and reinforcement training (MoIA) to improve explicit long-chain-of-thought generation of molecules.
Clark, Lionel Lon- don, Pablo Laguna, and Deirdre Shoemaker
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
citation-role summary
background 1
citation-polarity summary
fields
cs.CL 1years
2025 1verdicts
UNVERDICTED 1roles
background 1polarities
unclear 1representative citing papers
citing papers explorer
-
Mol-R1: Towards Explicit Long-CoT Reasoning in Molecule Discovery
Mol-R1 combines regulated in-context distillation (PRID) with iterative supervised and reinforcement training (MoIA) to improve explicit long-chain-of-thought generation of molecules.