LARC: Towards Human-level Constrained Retrosynthesis Planning through an Agentic Framework

TL;DR

LARC introduces an LLM-based agentic framework that effectively guides constrained retrosynthesis planning, achieving near-human success rates and outperforming baseline models by integrating agentic constraint evaluation directly into the planning process.

Contribution

This paper presents the first agentic LLM framework for constrained retrosynthesis, integrating tool-based constraint evaluation to improve planning success.

Findings

Achieves 72.9% success rate on constrained retrosynthesis tasks.

Outperforms baseline LLM models significantly.

Approaches human expert-level performance in planning efficiency.

Abstract

Large language model (LLM) agent evaluators leverage specialized tools to ground the rational decision-making of LLMs, making them well-suited to aid in scientific discoveries, such as constrained retrosynthesis planning. Constrained retrosynthesis planning is an essential, yet challenging, process within chemistry for identifying synthetic routes from commercially available starting materials to desired target molecules, subject to practical constraints. Here, we present LARC, the first LLM-based Agentic framework for Retrosynthesis planning under Constraints. LARC incorporates agentic constraint evaluation, through an Agent-as-a-Judge, directly into the retrosynthesis planning process, using agentic feedback grounded in tool-based reasoning to guide and constrain route generation. We rigorously evaluate LARC on a carefully curated set of 48 constrained retrosynthesis planning tasks across 3 constraint types. LARC achieves a 72.9% success rate on these tasks, vastly outperforming LLM baselines and approaching human expert-level success in substantially less time. The LARC framework is extensible, and serves as a first step towards an effective agentic tool or a co-scientist to human experts for constrained retrosynthesis.