Chemically Meaningful Textualization Enables Explainable Validation of Metal-Organic Frameworks by Large Language Models
This work provides an explainable and interpretable validation tool for researchers curating MOF databases, addressing the problem of chemically unreasonable or disordered structures that compromise simulation fidelity.
The authors developed a method to validate metal-organic framework (MOF) structures using large language models (LLMs) by converting crystallographic data into chemically meaningful text. They found that LLMs, when fine-tuned with specialized textual descriptors (mof2text), achieved performance comparable to graph-based models in identifying unreasonable MOFs and could generate diagnostic rationales for errors.
Computation-ready metal-organic framework (MOF) databases are essential for high-throughput screening, yet many reported crystal structures remain chemically unreasonable or disordered, compromising simulation fidelity. Existing validation approaches can identify non-computation-ready structures, but they often rely on heuristic rules, license requirement, or offer limited interpretability. Here, we show that large language models (LLMs) can serve as interpretable validators of MOF structures when crystallographic information is transformed into chemically meaningful text. By benchmarking nine descriptors, we find that successful LLM-based validation depends not on the amount of structural information alone, but on whether local coordination, framework connectivity, and chemical context are organized into a linguistically learnable representation. Fine-tuned LLMs using specialized descriptors (mof2text) achieve performance comparable to graph-based models in identifying unreasonable MOFs. Importantly, these models extend beyond black-box classification by generating diagnostic rationales for likely error sources, including abnormal bonding, connectivity, and charge states, as well as error-category predictions for annotated datasets. This work establishes chemically informed textualization as the key step that transforms LLMs from generic text models into practical and explainable tools for curating MOF databases.