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

arXiv: https://arxiv.org/abs/2509.23099 · HTML: https://mlsanity.com/paper/2509.23099/

## Key facts

- Cited by 4 papers in the map (data through 14 Oct 2026).
- Builds on 23 of its 51 references that are in the map.
- First posted on arXiv 27 Sep 2025.

## Builds on (most cited first, 23 of 23)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformer](https://mlsanity.com/paper/1910.10683/) (2019)
- [Scaling Laws for Neural Language Models](https://mlsanity.com/paper/2001.08361/) (2020)
- [Learning Phrase Representations using RNN Encoder-Decoder for Statistical Machine Translation](https://mlsanity.com/paper/1406.1078/) (2014)
- [Automatic chemical design using a data-driven continuous representation of molecules](https://mlsanity.com/paper/1610.02415/) (2016)
- [Equivariant Diffusion for Molecule Generation in 3D](https://mlsanity.com/paper/2203.17003/) (2022)
- [Junction Tree Variational Autoencoder for Molecular Graph Generation](https://mlsanity.com/paper/1802.04364/) (2018)
- [DiGress: Discrete Denoising diffusion for graph generation](https://mlsanity.com/paper/2209.14734/) (2022)
- [Molecular Sets (MOSES): A Benchmarking Platform for Molecular Generation Models](https://mlsanity.com/paper/1811.12823/) (2018)
- [Hierarchical Generation of Molecular Graphs using Structural Motifs](https://mlsanity.com/paper/2002.03230/) (2020)
- [Grammar-Constrained Decoding for Structured NLP Tasks without Finetuning](https://mlsanity.com/paper/2305.13971/) (2023)
- [Geometric Latent Diffusion Models for 3D Molecule Generation](https://mlsanity.com/paper/2305.01140/) (2023)
- [What can Large Language Models do in chemistry? A comprehensive benchmark on eight tasks](https://mlsanity.com/paper/2305.18365/) (2023)
- [GraphDF: A Discrete Flow Model for Molecular Graph Generation](https://mlsanity.com/paper/2102.01189/) (2021)
- [Syntax-Directed Variational Autoencoder for Structured Data](https://mlsanity.com/paper/1802.08786/) (2018)
- [Empowering Molecule Discovery for Molecule-Caption Translation with Large Language Models: A ChatGPT Perspective](https://mlsanity.com/paper/2306.06615/) (2023)
- [Keeping it Simple: Language Models can learn Complex Molecular Distributions](https://mlsanity.com/paper/2112.03041/) (2021)
- [Molecular Hypergraph Grammar with its Application to Molecular Optimization](https://mlsanity.com/paper/1809.02745/) (2018)
- [Group SELFIES: A Robust Fragment-Based Molecular String Representation](https://mlsanity.com/paper/2211.13322/) (2022)
- [Text-Guided Molecule Generation with Diffusion Language Model](https://mlsanity.com/paper/2402.13040/) (2024)
- [Data-Efficient Graph Grammar Learning for Molecular Generation](https://mlsanity.com/paper/2203.08031/) (2022)
- [Sketch-Guided Constrained Decoding for Boosting Blackbox Large Language Models without Logit Access](https://mlsanity.com/paper/2401.09967/) (2024)
- [Data-Efficient Molecular Generation with Hierarchical Textual Inversion](https://mlsanity.com/paper/2405.02845/) (2024)

## Built on by (most cited first, 4 of 4)

- [LLM-Guided Dynamic Action Spaces for Synthesizable Molecular Optimization](https://mlsanity.com/paper/2604.07669/) (2026)
- [SynCraft: Guiding Large Language Models to Predict Edit Sequences for Molecular Synthesizability Optimization](https://mlsanity.com/paper/2512.20333/) (2025)
- [How Creative Are Large Language Models in Generating Molecules?](https://mlsanity.com/paper/2604.18031/) (2026)
- [Agentic Molecular Recovery via Molecule-Aware Exploration](https://mlsanity.com/paper/2606.05847/) (2026)

## Abstract (arXiv)

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. Code is available at https://github.com/wentao228/SmiSelf.