# LLM-Guided Dynamic Action Spaces for Synthesizable Molecular Optimization

arXiv: https://arxiv.org/abs/2604.07669 · HTML: https://mlsanity.com/paper/2604.07669/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- Builds on 25 of its 39 references that are in the map.
- First posted on arXiv 9 Apr 2026.

## Builds on (most cited first, 25 of 25)

- [ReAct: Synergizing Reasoning and Acting in Language Models](https://mlsanity.com/paper/2210.03629/) (2022)
- [Efficient Memory Management for Large Language Model Serving with PagedAttention](https://mlsanity.com/paper/2309.06180/) (2023)
- [ChemCrow: Augmenting large-language models with chemistry tools](https://mlsanity.com/paper/2304.05376/) (2023)
- [Flow Network based Generative Models for Non-Iterative Diverse Candidate Generation](https://mlsanity.com/paper/2106.04399/) (2021)
- [Therapeutics Data Commons: Machine Learning Datasets and Tasks for Drug Discovery and Development](https://mlsanity.com/paper/2102.09548/) (2021)
- [What can Large Language Models do in chemistry? A comprehensive benchmark on eight tasks](https://mlsanity.com/paper/2305.18365/) (2023)
- [Sample Efficiency Matters: A Benchmark for Practical Molecular Optimization](https://mlsanity.com/paper/2206.12411/) (2022)
- [Mol-Instructions: A Large-Scale Biomolecular Instruction Dataset for Large Language Models](https://mlsanity.com/paper/2306.08018/) (2023)
- [ChemBO: Bayesian Optimization of Small Organic Molecules with Synthesizable Recommendations](https://mlsanity.com/paper/1908.01425/) (2019)
- [DrugAssist: A Large Language Model for Molecule Optimization](https://mlsanity.com/paper/2401.10334/) (2023)
- [Sample-efficient Multi-objective Molecular Optimization with GFlowNets](https://mlsanity.com/paper/2302.04040/) (2023)
- [RGFN: Synthesizable Molecular Generation Using GFlowNets](https://mlsanity.com/paper/2406.08506/) (2024)
- [SynFlowNet: Design of Diverse and Novel Molecules with Synthesis Constraints](https://mlsanity.com/paper/2405.01155/) (2024)
- [LLaMo: Large Language Model-based Molecular Graph Assistant](https://mlsanity.com/paper/2411.00871/) (2024)
- [Generative Flows on Synthetic Pathway for Drug Design](https://mlsanity.com/paper/2410.04542/) (2024)
- [Projecting Molecules into Synthesizable Chemical Spaces](https://mlsanity.com/paper/2406.04628/) (2024)
- [MT-Mol:Multi Agent System with Tool-based Reasoning for Molecular Optimization](https://mlsanity.com/paper/2505.20820/) (2025)
- [LDMol: A Text-to-Molecule Diffusion Model with Structurally Informative Latent Space Surpasses AR Models](https://mlsanity.com/paper/2405.17829/) (2024)
- [Deep Lead Optimization: Leveraging Generative AI for Structural Modification](https://mlsanity.com/paper/2404.19230/) (2024)
- [SynLlama: Generating Synthesizable Molecules and Their Analogs with Large Language Models](https://mlsanity.com/paper/2503.12602/) (2025)
- [mCLM: A Modular Chemical Language Model that Generates Functional and Makeable Molecules](https://mlsanity.com/paper/2505.12565/) (2025)
- [Exploring Synthesizable Chemical Space with Iterative Pathway Refinements](https://mlsanity.com/paper/2509.16084/) (2025)
- [Procedural Synthesis of Synthesizable Molecules](https://mlsanity.com/paper/2409.05873/) (2024)
- [An efficient graph generative model for navigating ultra-large combinatorial synthesis libraries](https://mlsanity.com/paper/2211.04468/) (2022)
- [How to Make Large Language Models Generate 100% Valid Molecules?](https://mlsanity.com/paper/2509.23099/) (2025)

## Built on by (most cited first, 2 of 2)

- [SMDD-Bench: Can LLMs Solve Real-World Small Molecule Drug Design Tasks?](https://mlsanity.com/paper/2605.21740/) (2026)
- [My Chemical Harness: Evolutionary Molecular Design over Synthetic Pathways with Large Language Model Agents](https://mlsanity.com/paper/2606.11256/) (2026)

## Abstract (arXiv)

Synthesizable molecular optimization seeks to improve target properties while ensuring that molecular modifications follow feasible synthetic pathways. Existing synthesis-aware methods typically rely on exploring a large space of candidate transformations defined by reaction templates and purchasable building blocks. This search becomes even more challenging when property improvement requires multiple reaction steps, as the space expands further along the pathway. To address this challenge, we introduce MolReAct, which reformulates molecular optimization as search over compact reaction spaces proposed by a tool-augmented large language model (LLM). At each step, the LLM combines its prior chemical knowledge with cheminformatics tools to identify a molecule-specific set of compatible reactions, preserving synthesizability while making multi-step optimization feasible. Given this compact action space, we further leverage Group Relative Policy Optimization (GRPO) with the terminal oracle reward to improve long-term decision-making over multiple reaction steps. Across diverse molecular optimization tasks, MolReAct achieves the highest Top-10 score on 11 of 14 tasks and the best sample efficiency on 12 of 14 tasks, outperforming existing baselines under limited oracle budgets. Beyond these gains, MolReAct also provides each optimized molecule with a template-grounded synthetic pathway.