ChemBOMAS: Accelerated BO in Chemistry with LLM-Enhanced Multi-Agent System

2025/09/10 by Dong Han, Zizheng Ai, Han, Dong +34 · 1 citation
Computer Science · Materials Science · #Baseline (sea) #Bayesian optimization #Bayesian probability #Branch and bound #Computational Drug Discovery Methods #FOS: Computer and information sciences #Initialization #Linear subspace #Machine Learning (cs.LG) #Machine Learning and Data Classification #Machine Learning in Materials Science #Set (abstract data type) #Upper and lower bounds

paper · pdf · doi:10.48550/arxiv.2509.08736

published in arXiv (Cornell University) (Cornell University)

openalex publication_date 2025/09/10 · openalex created_date 2025/10/10 · openalex updated_date 2026/08/05

Abstract

Bayesian optimization (BO) is a powerful tool for scientific discovery in chemistry, yet its efficiency is often hampered by the sparse experimental data and vast search space. Here, we introduce ChemBOMAS: a large language model (LLM)-enhanced multi-agent system that accelerates BO through synergistic data- and knowledge-driven strategies. Firstly, the data-driven strategy involves an 8B-scale LLM regressor fine-tuned on a mere 1% labeled samples for pseudo-data generation, robustly initializing the optimization process. Secondly, the knowledge-driven strategy employs a hybrid Retrieval-Augmented Generation approach to guide LLM in dividing the search space while mitigating LLM hallucinations. An Upper Confidence Bound algorithm then identifies high-potential subspaces within this established partition. Across the LLM-refined subspaces and supported by LLM-generated data, BO achieves the improvement of effectiveness and efficiency. Comprehensive evaluations across multiple scientific benchmarks demonstrate that ChemBOMAS set a new state-of-the-art, accelerating optimization efficiency by up to 5-fold compared to baseline methods.

Citations

• Monte Carlo Tree Search based Space Transfer for Black-box Optimization
• Analyzing and Reducing Catastrophic Forgetting in Parameter Efficient Tuning
• Large Language Models to Enhance Bayesian Optimization
• Identifying general reaction conditions by bandit optimization
• Recent Advances in Bayesian Optimization
• Bayesian reaction optimization as a tool for chemical synthesis
• Learning Search Space Partition for Black-box Optimization using Monte Carlo Tree Search
• Sample-Efficient Optimization in the Latent Space of Deep Generative Models via Weighted Retraining
• Sample-Efficient Neural Architecture Search by Learning Action Space
• Taking the Human Out of the Loop: A Review of Bayesian Optimization

Cited by

• AutoScientists: Self-Organizing Agent Teams for Long-Running Scientific Experimentation

Related

• Pre-trained knowledge elevates large language models beyond traditional chemical reaction optimizers
• A Sober Look at LLMs for Material Discovery: Are They Actually Good for Bayesian Optimization Over Molecules?
• HypBO: Accelerating Black-Box Scientific Experiments Using Experts' Hypotheses
• Reasoning BO: Enhancing Bayesian Optimization with Long-Context Reasoning Power of LLMs
• Distilling and exploiting quantitative insights from Large Language Models for enhanced Bayesian optimization of chemical reactions