RetroReasoner: A Reasoning LLM for Strategic Retrosynthesis Prediction

TL;DR

RetroReasoner is a novel LLM-based model that incorporates strategic reasoning and reinforcement learning to improve the accuracy and diversity of retrosynthesis predictions in organic chemistry.

Contribution

It introduces a new framework combining supervised fine-tuning with synthetic disconnection rationales and reinforcement learning with round-trip accuracy for better retrosynthesis prediction.

Findings

Outperforms prior models in accuracy and diversity

Generates more feasible and strategic reactant proposals

Handles challenging reaction cases effectively

Abstract

Retrosynthesis prediction is a core task in organic synthesis that aims to predict reactants for a given product molecule. Traditionally, chemists select a plausible bond disconnection and derive corresponding reactants, which is time-consuming and requires substantial expertise. While recent advancements in molecular large language models (LLMs) have made progress, many methods either predict reactants without strategic reasoning or conduct only a generic product analysis, rather than reason explicitly about bond-disconnection strategies that logically lead to the choice of specific reactants. To overcome these limitations, we propose RetroReasoner, a retrosynthetic reasoning model that leverages chemists' strategic thinking. RetroReasoner is trained using both supervised fine-tuning (SFT) and reinforcement learning (RL). For SFT, we introduce SyntheticRetro, a framework that generates structured disconnection rationales alongside reactant predictions. In the case of RL, we apply a round-trip accuracy as reward, where predicted reactants are passed through a forward synthesis model, and predictions are rewarded when the forward-predicted product matches the original input product. Experimental results show that RetroReasoner not only outperforms prior baselines but also generates a broader range of feasible reactant proposals, particularly in handling more challenging reaction instances.