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

arXiv: https://arxiv.org/abs/2505.20820 · HTML: https://mlsanity.com/paper/2505.20820/

## Key facts

- Cited by 15 papers in the map (data through 14 Oct 2026).
- Placed in Clinical and biomedical applications, one of 17 directions in LLM reasoning.
- Builds on 5 of its 38 references that are in the map.
- First posted on arXiv 27 May 2025.

## Placed in

- LLM reasoning / [Clinical and biomedical applications](https://mlsanity.com/lr/clinical-and-biomedical-applications/)
- LLM agents and tool use / [Medical and scientific domain agents](https://mlsanity.com/agents/medical-and-scientific-domain-agents/)

## Builds on (most cited first, 5 of 5)

- [Evaluating Large Language Models Trained on Code](https://mlsanity.com/paper/2107.03374/) (2021)
- [OPT: Open Pre-trained Transformer Language Models](https://mlsanity.com/paper/2205.01068/) (2022)
- [DeepSeek-V3 Technical Report](https://mlsanity.com/paper/2412.19437/) (2024)
- [Therapeutics Data Commons: Machine Learning Datasets and Tasks for Drug Discovery and Development](https://mlsanity.com/paper/2102.09548/) (2021)
- [Investigating Fairness Disparities in Peer Review: A Language Model Enhanced Approach](https://mlsanity.com/paper/2211.06398/) (2022)

## Built on by (most cited first, 15 of 15)

- [From AI for Science to Agentic Science: A Survey on Autonomous Scientific Discovery](https://mlsanity.com/paper/2508.14111/) (2025)
- [El Agente Quntur: A research collaborator agent for quantum chemistry](https://mlsanity.com/paper/2602.04850/) (2026)
- [Exploring Modularity of Agentic Systems for Drug Discovery](https://mlsanity.com/paper/2506.22189/) (2025)
- [LLM-Guided Dynamic Action Spaces for Synthesizable Molecular Optimization](https://mlsanity.com/paper/2604.07669/) (2026)
- [Mol-Debate: Multi-Agent Debate Improves Structural Reasoning in Molecular Design](https://mlsanity.com/paper/2604.20254/) (2026)
- [PepEVOLVE: Position-Aware Dynamic Peptide Optimization via Group-Relative Advantage](https://mlsanity.com/paper/2511.16912/) (2025)
- [INDIBATOR: Diverse and Fact-Grounded Individuality for Multi-Agent Debate in Molecular Discovery](https://mlsanity.com/paper/2602.01815/) (2026)
- [MolLIBRA: Genetic Molecular Optimization with Multi-Fingerprint Surrogates and Text-Molecule Aligned Critic](https://mlsanity.com/paper/2602.07002/) (2026)
- [Deciphering Scientific Reasoning Steps from Outcome Data for Molecule Optimization](https://mlsanity.com/paper/2603.20262/) (2026)
- [MolDeTox: Evaluating Language Model's Stepwise Fragment Editing for Molecular Detoxification](https://mlsanity.com/paper/2605.12181/) (2026)
- [Agentic Molecular Recovery via Molecule-Aware Exploration](https://mlsanity.com/paper/2606.05847/) (2026)
- [Molecular LLM Agents: From Architectural Design to Scientific Autonomy](https://mlsanity.com/paper/2608.23104/) (2026)
- [M3OS: A Monte Carlo Graph Search-Orchestrated Multi-Agent LLM System for Evidence-Traced Molecular Optimization](https://mlsanity.com/paper/2609.34491/) (2026)
- [From Benchmark to Bench: Can Agents Survive Real-World Drug Discovery?](https://mlsanity.com/paper/2610.06411/) (2026)
- [Masked Diffusion Language Agents for Tool-Integrated Chemical Reasoning](https://mlsanity.com/paper/2610.13330/) (2026)

## Abstract (arXiv)

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 MT-Mol, 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. MT-Mol produces molecules with tool-aligned and stepwise reasoning through the interaction between the analyst agents, a molecule-generating scientist, a reasoning-output verifier, and a reviewer agent. 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.