RetroAgent: Harnessing LLMs to Search Over Structured Memory for Agentic Retrosynthesis Planning
For chemists and drug discovery, this work improves retrosynthesis planning by enabling LLMs to reason over full search states, but the gains are incremental over existing tree-search methods.
RetroAgent introduces an LLM agent that uses structured memory to improve multi-step retrosynthesis planning, achieving strong performance and generalization on in-distribution and out-of-distribution benchmarks.
Multi-step retrosynthesis planning seeks to decompose a target molecule into commercially available building blocks through a sequence of feasible reactions. The vast combinatorial search space makes this task challenging even for expert chemists. Traditional methods combine tree search with offline-trained value networks that score candidates in isolation, without reasoning about complete multi-step routes. Recent work leverages Large Language Models (LLMs) for this task, but relies on simple interfaces that limit exploration of the full search space. We introduce RetroAgent, an LLM agent that bridges symbolic search and neural reasoning through a harness with structured memory. Through memory and chemistry tools, the agent observes the full search state, including explored routes, available alternatives, and properties of intermediates, enabling informed decisions grounded in both global progress and domain knowledge. Experiments on in-distribution and out-of-distribution benchmarks demonstrate that RetroAgent delivers strong performance and generalization.