Metraipolis › archive › dc_4a022128bb23

Chemically Meaningful Textualization Enables Explainable Validation of Metal-Organic Frameworks by Large Language Models — arXiv:2608.11283

Source
arXiv (en)
Original
https://arxiv.org/abs/2608.11283
License (as archived)
CC0-1.0
Archived
2026-10-19T05:15:44.127Z
SHA-256 of the archived text
eb663b71673312a5c4ea024714800024abcbc8edf93acd01782fe9df133aadc2
Stable URL of this page
https://metraipolis.com/documento/dc_4a022128bb23

How to cite

Chemically Meaningful Textualization Enables Explainable Validation of Metal-Organic Frameworks by Large Language Models — arXiv:2608.11283 — arXiv (en), https://arxiv.org/abs/2608.11283 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_4a022128bb23 (Metraipolis archive dc_4a022128bb23, sha256 eb663b71673312a5c4ea024714800024abcbc8edf93acd01782fe9df133aadc2)
BibTeX
@misc{dc4a022128bb23,
  title = {Chemically Meaningful Textualization Enables Explainable Validation of Metal-Organic Frameworks by Large Language Models — arXiv:2608.11283},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2608.11283},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_4a022128bb23; sha256 eb663b71673312a5c4ea024714800024abcbc8edf93acd01782fe9df133aadc2},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.11283. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

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.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-08-13 in cs.AI).
Categorie arXiv: Materials Science (cond-mat.mtrl-sci) ; Artificial Intelligence (cs.AI).
Pagina del lavoro: https://arxiv.org/abs/2608.11283 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito).
Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina.
Letto il 2026-10-19T05:08:46.673Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-08-13?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).