MT-Mol: Multi Agent System with Tool-based Reasoning for Molecular Optimization

ACL ARR 2025 May Submission6580 Authors

Abstract: Large language models (LLMs) have large potential for molecular optimization, as they can gather external chemistry tools and enable collaborative interactions to iteratively refine molecular candidates. However, this potential remains underexplored, particularly in the context of structured reasoning, interpretability, and comprehensive tool-grounded molecular optimization. To address this gap, we introduce \methodname, a multi-agent framework for molecular optimization that leverages tool-guided reasoning and role-specialized LLM agents. Our system incorporates comprehensive RDKit tools, categorized into five distinct domains: structural descriptors, electronic and topological features, fragment-based functional groups, molecular representations, and miscellaneous chemical properties. Each category is managed by an expert analyst agent, responsible for extracting task-relevant tools and enabling interpretable, chemically grounded feedback. Through the interaction between the analyst agents, a molecule-generating scientist, a reasoning-output verifier, and a reviewer agent produce molecules with tool-aligned and stepwise reasoning. As a result, we show that our framework shows the state-of-the-art performance of the PMO-1K benchmark on 17 out of 23 tasks.

Paper Type: Long

Research Area: Summarization

Research Area Keywords: healthcare applications

Contribution Types: Model analysis & interpretability, NLP engineering experiment

Languages Studied: English

Keywords: healthcare applications

Submission Number: 6580