Training Chemical Plausibility-Aware Large Language Models for Single-Step Retrosynthesis
Single-step retrosynthesis is a central component of computer-aided synthesis planning, yet its intrinsically one-to-many nature is poorly captured by single-answer evaluation and benchmarking protocols. To address this, we introduce Top-K prompting as a robust training and inference paradigm to better capture diverse, plausible reaction predictions. We compile CREED-CCV-2+USPTO-XL, an ultra-large-scale dataset of ~45.6 million verified reactions to train the C3LM (Chemistry Constraint-Consistent Language Model). By integrating fine-tuning with ChemCensor-based and novelty-oriented rewards, ou
Lineage graph
Paper → model → repo connections mined from source citations (Tier-1 exact match).
Why these links exist
Every edge carries a method, confidence, and the source snippet that justified it — so bad links are debuggable.
- LinkedLinked via arxiv author · 85%Bogdan Zagribelnyy →
“Training Chemical Plausibility-Aware Large Language Models for Single-Step Retrosynthesis”
- LinkedLinked via arxiv author · 85%Ivan Ilin →
“Training Chemical Plausibility-Aware Large Language Models for Single-Step Retrosynthesis”
- LinkedLinked via arxiv author · 85%Nikita Bondarev →
“Training Chemical Plausibility-Aware Large Language Models for Single-Step Retrosynthesis”
- LinkedLinked via arxiv author · 85%Maksim Kuznetsov →
“Training Chemical Plausibility-Aware Large Language Models for Single-Step Retrosynthesis”
- LinkedLinked via arxiv author · 85%Mathieu Reymond →
“Training Chemical Plausibility-Aware Large Language Models for Single-Step Retrosynthesis”
- LinkedLinked via arxiv author · 85%Vladimir Aladinskiy →
“Training Chemical Plausibility-Aware Large Language Models for Single-Step Retrosynthesis”
- LinkedLinked via arxiv author · 85%Alex Aliper →
“Training Chemical Plausibility-Aware Large Language Models for Single-Step Retrosynthesis”
- LinkedLinked via arxiv author · 85%Alex Zhavoronkov →
“Training Chemical Plausibility-Aware Large Language Models for Single-Step Retrosynthesis”