# Closed-Loop Molecular Inverse Design with Semantic LLM Surrogates

arXiv: https://arxiv.org/abs/2608.22967 · HTML: https://mlsanity.com/paper/2608.22967/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 27 of its 49 references that are in the map.
- First posted on arXiv 24 Aug 2026.

## Builds on (most cited first, 27 of 27)

- [The Llama 3 Herd of Models](https://mlsanity.com/paper/2407.21783/) (2024)
- [Mistral 7B](https://mlsanity.com/paper/2310.06825/) (2023)
- [Qwen2 Technical Report](https://mlsanity.com/paper/2407.10671/) (2024)
- [Large Language Models as Optimizers](https://mlsanity.com/paper/2309.03409/) (2023)
- [Automatic chemical design using a data-driven continuous representation of molecules](https://mlsanity.com/paper/1610.02415/) (2016)
- [A Tutorial on Bayesian Optimization](https://mlsanity.com/paper/1807.02811/) (2018)
- [Galactica: A Large Language Model for Science](https://mlsanity.com/paper/2211.09085/) (2022)
- [Junction Tree Variational Autoencoder for Molecular Graph Generation](https://mlsanity.com/paper/1802.04364/) (2018)
- [Scalable Global Optimization via Local Bayesian Optimization](https://mlsanity.com/paper/1910.01739/) (2019)
- [Graph Convolutional Policy Network for Goal-Directed Molecular Graph Generation](https://mlsanity.com/paper/1806.02473/) (2018)
- [ChemBERTa: Large-Scale Self-Supervised Pretraining for Molecular Property Prediction](https://mlsanity.com/paper/2010.09885/) (2020)
- [GuacaMol: Benchmarking Models for De Novo Molecular Design](https://mlsanity.com/paper/1811.09621/) (2018)
- [Grammar Variational Autoencoder](https://mlsanity.com/paper/1703.01925/) (2017)
- [Molecular Sets (MOSES): A Benchmarking Platform for Molecular Generation Models](https://mlsanity.com/paper/1811.12823/) (2018)
- [Molecular De Novo Design through Deep Reinforcement Learning](https://mlsanity.com/paper/1704.07555/) (2017)
- [Therapeutics Data Commons: Machine Learning Datasets and Tasks for Drug Discovery and Development](https://mlsanity.com/paper/2102.09548/) (2021)
- [Bayesian Optimization in a Billion Dimensions via Random Embeddings](https://mlsanity.com/paper/1301.1942/) (2013)
- [Translation between Molecules and Natural Language](https://mlsanity.com/paper/2204.11817/) (2022)
- [Multi-modal Molecule Structure-text Model for Text-based Retrieval and Editing](https://mlsanity.com/paper/2212.10789/) (2022)
- [Deep Reinforcement Learning for De-Novo Drug Design](https://mlsanity.com/paper/1711.10907/) (2017)
- [Sample Efficiency Matters: A Benchmark for Practical Molecular Optimization](https://mlsanity.com/paper/2206.12411/) (2022)
- [Unexpected Improvements to Expected Improvement for Bayesian Optimization](https://mlsanity.com/paper/2310.20708/) (2023)
- [Large Language Models to Enhance Bayesian Optimization](https://mlsanity.com/paper/2402.03921/) (2024)
- [A survey on high-dimensional Gaussian process modeling with application to Bayesian optimization](https://mlsanity.com/paper/2111.05040/) (2021)
- [Multimodal Large Language Models for Inverse Molecular Design with Retrosynthetic Planning](https://mlsanity.com/paper/2410.04223/) (2024)
- [A Survey of Large Language Models for Text-Guided Molecular Discovery: from Molecule Generation to Optimization](https://mlsanity.com/paper/2505.16094/) (2025)
- [AOT*: Efficient Synthesis Planning via LLM-Empowered AND-OR Tree Search](https://mlsanity.com/paper/2509.20988/) (2025)

## Built on by (most cited first, 1 of 1)

- [LLM-to-Flow Distillation: Teaching Natural Language Goals to Scientific Generators](https://mlsanity.com/paper/2610.17246/) (2026)

## Abstract (arXiv)

Practical molecular inverse design is rarely a one-shot generation problem; it often involves enriching a pool of molecules that match a desired property profile under a limited oracle budget. Bayesian optimization (BO) offers a natural framework for this setting, with standard Gaussian-process (GP) surrogates operating in continuous embedding spaces. We propose LoopMol, a closed-loop framework whose large language model (LLM) surrogate reasons directly over task instructions, SMILES-level optimization histories, and oracle feedback. At each iteration, the surrogate selects informative reference molecules under an exploration--exploitation principle, optionally with a concise guidance sentence. These outputs condition the next generation round and yield an inspectable optimization trace. Experiments on six drug and material tasks across three Llamole backbones show that the LLM-based variant improves over one-shot generation and is competitive with or stronger than GP-based BO. They also reveal a domain-dependent interface: reference-only conditioning performs better on binary drug tasks, while adding a concise guidance sentence generally benefits continuous material tasks.