Training Chemical Plausibility-Aware Large Language Models for Single-Step Retrosynthesis

2026-08-19

AI Summary

为解决单步逆合成预测多样性不足的问题，通过Top-K提示法和大规模数据集训练化学合理性感知的语言模型，结合精细调整与奖励机制提高性能。

Abstract

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, our model achieves state-of-the-art performance on the OOD URSA-expert-2026 benchmark. Further analysis of reaction uniqueness shows that LLMs and conventional models explore complementary reaction spaces, motivating ensemble-based retrosynthesis systems. Overall, our results establish Top-K, plausibility-aware training as a practical new direction for robust future LLM-based synthesis planning.

Problem

• single-step retrosynthesis
• one-to-many nature
• single-answer evaluation

Innovation

• Top-K prompting
• plausibility-aware training
• ChemCensor-based rewards
• novelty-oriented rewards
• ensemble-based retrosynthesis systems