# Unleashing LLMs in Bayesian Optimization: Preference-Guided Framework for Scientific Discovery

arXiv: https://arxiv.org/abs/2605.17976 · HTML: https://mlsanity.com/paper/2605.17976/

## Key facts

- Cited by 7 papers in the map (data through 14 Oct 2026).
- Builds on 12 of its 36 references that are in the map.
- First posted on arXiv 18 May 2026.

## Builds on (most cited first, 12 of 12)

- [On Kernelized Multi-armed Bandits](https://mlsanity.com/paper/1704.00445/) (2017)
- [Large Language Models to Enhance Bayesian Optimization](https://mlsanity.com/paper/2402.03921/) (2024)
- [Learning Search Space Partition for Black-box Optimization using Monte Carlo Tree Search](https://mlsanity.com/paper/2007.00708/) (2020)
- [Preferential Bayesian Optimization](https://mlsanity.com/paper/1704.03651/) (2017)
- [ADO-LLM: Analog Design Bayesian Optimization with In-Context Learning of Large Language Models](https://mlsanity.com/paper/2406.18770/) (2024)
- [Olympus: a benchmarking framework for noisy optimization and experiment planning](https://mlsanity.com/paper/2010.04153/) (2020)
- [A General Framework for User-Guided Bayesian Optimization](https://mlsanity.com/paper/2311.14645/) (2023)
- [No-regret Algorithms for Multi-task Bayesian Optimization](https://mlsanity.com/paper/2008.08885/) (2020)
- [Adaptive Kernel Design for Bayesian Optimization Is a Piece of CAKE with LLMs](https://mlsanity.com/paper/2509.17998/) (2025)
- [LLINBO: Trustworthy LLM-in-the-Loop Bayesian Optimization](https://mlsanity.com/paper/2505.14756/) (2025)
- [Reasoning BO: Enhancing Bayesian Optimization with Long-Context Reasoning Power of LLMs](https://mlsanity.com/paper/2505.12833/) (2025)
- [Bayesian Optimization Augmented with Actively Elicited Expert Knowledge](https://mlsanity.com/paper/2208.08742/) (2022)

## Built on by (most cited first, 7 of 7)

- [LABO: LLM-Accelerated Bayesian Optimization through Broad Exploration and Selective Experimentation](https://mlsanity.com/paper/2605.22054/) (2026)
- [Evidence-Gated LLM Priors for Multi-Objective Bayesian Optimization](https://mlsanity.com/paper/2606.01730/) (2026)
- [Compressing the Validation Bottleneck: An Agentic Self-Driving Lab for Scientific Discovery](https://mlsanity.com/paper/2607.04508/) (2026)
- [Bayesian Optimization with Rich Auxiliary Information via LLMs](https://mlsanity.com/paper/2609.19437/) (2026)
- [Harnessing Large Language Models to Compile Task-Relevant Context into Bayesian Optimisation](https://mlsanity.com/paper/2609.36788/) (2026)
- [LLM-to-Flow Distillation: Teaching Natural Language Goals to Scientific Generators](https://mlsanity.com/paper/2610.17246/) (2026)
- [Agentic-MatriBO: Capability-Compiled Agentic Bayesian Optimization for Materials and Chemical Design](https://mlsanity.com/paper/2610.17832/) (2026)

## Abstract (arXiv)

Scientific discovery is increasingly constrained by costly experiments and limited resources, underscoring the need for efficient optimization in AI for science. Bayesian Optimization (BO), though widely adopted for balancing exploration and exploitation, often exhibits slow cold-start performance and poor scalability in high-dimensional settings, limiting its applicability in real-world scientific problems. To overcome these challenges, we propose LLM-Guided Bayesian Optimization (LGBO), the first LLM preference-guided BO framework that continuously integrates the semantic reasoning of large language models (LLMs) into the optimization loop. Unlike prior works that use LLMs only for warm-start initialization or candidate generation, LGBO introduces a region-lifted preference mechanism that embeds LLM-driven preferences into every iteration, shifting the surrogate mean in a stable and controllable way. Theoretically, we prove that LGBO does not perform significantly worse than standard BO in the worst case, while achieving significantly faster convergence when preferences align with the objective. Empirically, LGBO consistently outperforms existing methods across diverse dry benchmarks in physics, chemistry, biology, and materials science. Most notably, in a new wet-lab optimization of Fe-Cr battery electrolytes, LGBO attains 90% of the best observed value within 6 iterations, whereas standard BO and existing LLM-augmented baselines require more than 10. Together, these results suggest that LGBO offers a promising direction for integrating LLMs into scientific optimization workflows.