How to Make Large Language Models Generate 100% Valid Molecules?

Wen Tao, Jing Tang, Alvin Chan, Bryan Hooi, Baolong Bi, Nanyun Peng, Yuansheng Liu, and Yiwei Wang, in Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing (EMNLP), 2025.

Download the full text

Abstract

Molecule generation is key to drug discovery and materials science, enabling the design of novel compounds with specific properties. Large language models (LLMs) can learn to perform a wide range of tasks from just a few examples. However, generating valid molecules using representations like SMILES is challenging for LLMs in few-shot settings. In this work, we explore how LLMs can generate 100% valid molecules. We evaluate whether LLMs can use SELFIES, a representation where every string corresponds to a valid molecule, for valid molecule generation but find that LLMs perform worse with SELFIES than with SMILES. We then examine LLMs’ ability to correct invalid SMILES and find their capacity limited. Finally, we introduce SmiSelf, a cross-chemical language framework for invalid SMILES correction. SmiSelf converts invalid SMILES to SELFIES using grammatical rules, leveraging SELFIES’ mechanisms to correct the invalid SMILES. Experiments show that SmiSelf ensures 100% validity while preserving molecular characteristics and maintaining or even enhancing performance on other metrics. SmiSelf helps expand LLMs’ practical applications in biomedicine and is compatible with all SMILES-based generative models.

Bib Entry

@inproceedings{tao2025smiself,
  title = {How to Make Large Language Models Generate 100\% Valid Molecules?},
  author = {Tao, Wen and Tang, Jing and Chan, Alvin and Hooi, Bryan and Bi, Baolong and Peng, Nanyun and Liu, Yuansheng and Wang, Yiwei},
  year = {2025},
  booktitle = {Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing (EMNLP)}
}