Retro-Expert: Collaborative Reasoning for Interpretable Retrosynthesis

TL;DR

Retro-Expert addresses interpretability gaps in retrosynthesis by integrating specialized models to build a chemical decision space and an LLM to perform knowledge-grounded reasoning, optimized via KGPO. The framework yields not only reactant predictions but also interpretable natural-language reasoning paths, bridging AI outputs with chemical principles. Empirical results on USPTO-50K/USPTO-FULL show superior accuracy over both LLM-based and non-LLM baselines and strong generalization, including an OOD ChemBench improvement. Wet-lab validation and PaRoutes experiments demonstrate practical utility and scalable reasoning.

Abstract

Retrosynthesis prediction aims to infer the reactant molecule based on a given product molecule, which is a fundamental task in chemical synthesis. However, existing models rely on static pattern-matching paradigm, which limits their ability to perform effective logic decision-making, leading to black-box decision-making. Building on this, we propose Retro-Expert, an interpretable retrosynthesis framework that performs collaborative reasoning by combining the complementary reasoning strengths of Large Language Models and specialized models via reinforcement learning. It outputs natural language explanations grounded in chemical logic through three components: (1) specialized models analyze the product to construct high-quality chemical decision space, (2) LLM-driven critical reasoning to generate predictions and corresponding interpretable reasoning path, and (3) reinforcement learning optimizing interpretable decision policy. Experiments show that Retro-Expert not only surpasses both LLM-based and specialized models across different metrics but also provides expert-aligned explanations that bridge the gap between AI predictions and actionable chemical insights.

Figures (21)

• Figure 1: Comparison between the conventional RL pipeline and our methods. Conventional methods require SFT as a necessary step for effective RL, whereas our methods directly optimizes the LLM by leveraging the decision space.
• Figure 2: Overview of the Retro-Expert. (1) Decision Space Construction: Specialized models first analyze the target product to construct a high-dimensional chemical decision space composed of high-quality candidate pathways. (2) LLM-driven Navigation: Then, the LLM, acting as a reasoning engine, strategically navigates this space. This involves a critical-generative process where it can either select the best candidate pathway or generate a novel one if all provided options are deemed inadequate. (3) Policy Optimization: Finally, this navigation policy is trained end-to-end via Knowledge-Guided Policy Optimization (KGPO) to ensure the resulting reasoning is both accurate and chemically sound. Please zoom in for a better view of the details.
• Figure 3: Comparison of correct predictions distributions across different models and Success rate of Retro-Expert’s self-correction mechanism.
• Figure 4: Visualization of the novel reaction pathways proposed by Retro-Expert and validated through wet-lab experiments. The details are provided in the appendix.
• Figure 5: Visualization of the KGPO's optimization pipeline. Our target is to generate the final answer with an interpretable reasoning pathway by reasoning within the decision space. Thus we further implement a multi-stage reward mechanism that the reasoning text. The mechanism guides the model in two ways: it encourages rejecting incorrect candidates (e.g., "alkylation reaction would not" receives a reward of 1) while promoting inference of the correct reaction type (e.g., "deprotection reaction" receives a reward of 1).