MOFjson: An Explicit Topological Representation of Metal-Organic Frameworks for Language Models

Anabel Yong; Matthew Hart; Philippe Schwaller

cs.LGOct 25, 2026

MOFjson: An Explicit Topological Representation of Metal-Organic Frameworks for Language Models

Authors: Anabel Yong, Matthew Hart, Philippe Schwaller

Organizations: Laboratory of Artificial Chemical Intelligence (LIAC), EPFL & National Centre of Competence in Research (NCCR) Separations

Abstract

Language models for metal-organic frameworks (MOFs) rely on formats that separate atomic geometry from structural topology, forcing models to choose between unstructured coordinate lists and non-spatial string identifiers[1]. We present MOFjson, the first representation separating structural data into a raw atomic layer with explicit bond shift vectors and a derived layer exposing secondary building units (metal nodes and organic linkers) and network topology. Additionally, we introduce a benchmark with crystal-derived ground truth across four question categories. Across four models, MOFjson beats all three baselines in every category: 100% vs. 18-47% on reaching a block's nearest periodic copy, 96% vs. 71% on crystal system characterisation, and 60% vs. 48% on metal node and organic linker coordination. Ablating shift labels drops accuracy to 12.7%, proving models read the periodic graph rather than the surface format. MOFjson gives language models structure to read rather than reconstruct, establishing a foundation for agentic tool workflows and constrained MOF generation.

Figures & tables

Feature / Property	CIF	MOFid	MOFkey	MOFjson
Exact atomic positions	✓	✗	✗	✓
Explicit bond list	✗	✗	✗	✓
Node/linker identity	✗	✓	✓	✓
Net topology symbol	✗	✓	✓	✓
Boundary-crossing bonds	✗	✗	✗	✓ ( shift )
Reconstructible to full 3D structure	✓ (native)	✗	✗	✓ (verified)

Table 1: Existing MOF formats trade atomic detail for topology; MOFjson preserves both.

Figure 1: From crystal, to labelled quotient graph, to MOFjson, shown on the dia net. Net topology abstracts periodic structures into mathematical graphs of nodes and edges to define 3D connectivity [ 27 ] . Here the building blocks are vertices A and B. Left: a crystal fragment, with A bonded to four copies of B in different cells. Middle: the same fragment as a labelled quotient graph, one vertex per building block and one arc per bond with its integer shift vector. Right: MOFjson stores that graph directly, with building_blocks as vertices and net_edges as arcs (attachment points omitted). The dia net [ 28 ] denotes the 4-connected tetrahedral diamond topology.

Category	Example prompt
Symmetry	“In this metal–organic framework {representation}, what crystal system does the structure belong to?”
Chemical identity	“In this MOF {representation}, what metal(s) make up the SBU?” (or) “… what is the SMILES representation of the organic linker?”
Connectivity	“What is the connectivity of the Co-containing SBU – how many organic linkers coordinate directly to it in this MOF {representation}?”
Net path	“In this MOF {representation}, how few connections are needed to reach the nearest periodic image of this same Al-containing building unit?”

Table 2: One example prompt per QnA category in the benchmark. {representation} marks where the structure data (CIF, MOFjson, MOFid, or MOFkey) is inserted.

Figure 2: MOFjson yields the highest and most consistent accuracy of any representation tested. Left: accuracy by question category, each point one model’s mean over three runs; MOFjson leads on Symmetry, Net path, and Connectivity, the three requiring structural reasoning, but not Chemical identity, readable from atom labels in any representation. Right: grouped by model; MOFjson gives three of four models their best score. Full per-model, per-representation accuracy is reported in Appendix A.4

Question Category	Best Representation	Best Model	Accuracy (% $\pm$ SD)
Symmetry	MOFjson	gemini-3.7-flash	96.0 $\pm$ 4.0
Net path	MOFjson	gemini-3.7-flash	100.0 $\pm$ 0.0
Connectivity	MOFjson	gpt-5.6-luna	60.0 $\pm$ 4.0
Chemical identity	MOFjson	apertus-v1.5-70b	65.3 $\pm$ 8.1

Table 3: Best-performing (representation, model) pair for each question category.

Figure 3: Impact of periodic shifts on Net path accuracy. Performance of gemini-3.7-flash (

N=3

runs). Zeroing edge shift vectors causes performance to drop from

100.0\%

12.7\%\pm 6.4\%

Appendix figures & tables7 assets

Supplementary material from the paper’s appendix.

Appendix

Figure 4: MOFjson makes HKUST-1’s periodic graph explicit, where MOFid and MOFkey compress it into an opaque code. Left: HKUST-1’s MOFjson, showing the raw layer (lattice, atoms, bonds from CIF files) and the derived layer computed from it (building_blocks, net_edges, topology). Right: the same structure rendered in iRASPA software [ 40 ] , alongside its MOFid and MOFkey strings. Both identifiers encode HKUST-1’s net symbol tbo .

Figure 5: MOFjson conversion is lossless across the full CoRE MOF size range. Each of the 2,737 computation-ready CIFs [ 34 ] is parsed with pymatgen [ 20 ] , converted to the MOFjson raw layer, rebuilt into a structure from that layer alone, and compared against the source with StructureMatcher , which tests lattice parameters and site positions under symmetry operations rather than requiring byte-identical files. All 2,737 structures are recovered, from under 50 to over 800 atoms per unit cell (

n

above each bar), so fidelity does not degrade with cell size. Only the raw layer is tested: the derived fields are computed from it and are not required for reconstruction, and a building-block partition is not uniquely invertible in any case.

Model	Best Representation	Question Category	Accuracy (% $\pm$ SD)
claude-haiku-4.5	CIF	Symmetry	71.3 $\pm$ 7.6
gpt-5.6-luna	MOFjson	Net path	95.3 $\pm$ 4.2
gemini-3.7-flash	MOFjson	Net path	100.0 $\pm$ 0.0
apertus-v1.5-70b	MOFjson	Chemical identity	65.3 $\pm$ 8.1

Appendix

Table 4: Best-performing (representation, category) pair for each model, corresponding to the starred point in Figure 2 (right). Accuracy is reported as mean

\pm

SD across three runs.

Figure 6: MOFjson is consistently strong on symmetry questions. It matches or leads CIF in three models, outperforming it in gpt-5.6-luna (88.0% vs. 18.0%) and gemini-3.7-flash (96.0% vs. 68.7%), while statistically tying in claude-haiku-4.5 (70.0% vs. 71.3%). Because MOFid and MOFkey lack lattice geometry, both consistently lag behind. The sole exception is apertus-v1.5-70b , where performance across all representations collapses into a noisy 8–23% band (22.7% for MOFjson vs. 20.0% for CIF). Thus, while MOFjson’s lead over CIF depends on model capability, its edge over compressed string identifiers holds almost universally.

Figure 8: Chemical identity shows no consistent MOFjson advantage. For claude-haiku-4.5 , gpt-5.6-luna and gemini-3.7-flash , all four representations are statistically indistinguishable, clustered within 30–34% regardless of format. Only apertus-v1.5-70b separates, with MOFjson at 65.3% against 24.0–30.7% for the baselines. Because this appears in one model and nowhere else in the data results, we report it as a single-model observation rather than evidence of a representation effect, and treat Chemical identity as the category where the choice of representation matters least.

Question Category	Baseline	$n$	Cliff’s $\delta$	$p$ (Holm-adjusted)
Symmetry	CIF	588	0.25	$<0.001$
Symmetry	MOFid	588	0.39	$<0.001$
Symmetry	MOFkey	588	0.45	$<0.001$
Net path	CIF	580	0.29	$<0.001$
Net path	MOFid	580	0.28	$<0.001$
Net path	MOFkey	580	0.34	$<0.001$

Appendix

Table 5: Paired Wilcoxon signed-rank test comparing MOFjson against each baseline representation, per question category. Every run draws 50 structures and renders each of them in all 4 representations, so a structure contributes one matched pair per model for each MOFjson-baseline comparison. The 3 runs draw independently, so a structure can recur across runs; de-duplicating the 150 draws leaves 145–148 distinct structures per category (2–5 repeats of the same MOF structure), and each contributes one paired observation per model, giving

n=4\times

that count. Cliff’s

\delta

(range

-1

1

; positive favours MOFjson) is the probability MOFjson scores higher on a given structure minus the probability it scores lower.

p

-values are Holm-Bonferroni corrected across all 12 tests. MOFjson’s advantage is significant (

p<0.001

) against every baseline in every category.

Figure 10: MOFjson’s token cost is representation-efficient except for symmetry queries. Measured in input tokens, MOFjson uses 6–58

\times

fewer input tokens than CIF, approaching the compactness of MOFid and MOFkey. For symmetry, however, MOFjson requires the full per-atom coordinate list and exceeds even CIF’s token cost, reflecting the overhead of its verbose JSON atom encoding relative to CIF’s compact tabular atom-site format.

Materials Science

Materials science studies how composition, structure, processing, and defects determine material properties and how to design useful materials.

See all 306 papers

Chemically Meaningful Textualization Enables Explainable Validation of Metal-Organic Frameworks by Large Language Models
Aug 11, 2026Guobin Zhao, Xiao-Yan LiLLM Evaluation LLM Interpretability
Department of Chemistry, National University of Singapore, 3 Science Drive 3, Singapore 117543, Singapore
Interpretable Inverse Design of Metal-Organic Frameworks with Large Language Model Agents
Jun 28, 2026Kyungmin Nam, Seunghee Han, Jihan KimLarge Language Model-Guided Optimization Large Language Model-Guided Scientific Discovery
Department of Chemical and Biomolecular Engineering, Korea Advanced Institute of Science and Technology, Daejeon 34141, Republic of Korea.
MOF-VERIFY: A Failure-Aware Agentic Harness for MOF Hypothesis Verification
Oct 2, 2026Donghyun Lee, Taehoon Lee, Geonhee Ahn +10Scientific Reasoning in Language Models LLM Reliability
Institute for Multiscale Matters and System (IMMS), Ewha Womans University, Seoul, Republic of Korea · Lymeric, Seoul, Republic of Korea

Aug 11, 2026J/K move · Enter open · S save

Chemically Meaningful Textualization Enables Explainable Validation of Metal-Organic Frameworks by Large Language Models

Guobin Zhao, Xiao-Yan Li

Department of Chemistry, National University of Singapore, 3 Science Drive 3, Singapore 117543, Singapore

LLM Evaluation LLM Interpretability Materials Science Molecular Representation Learning

Open paper AskPDF arXiv

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.

Abstract

Materials Science

Explore similar work

Chemically Meaningful Textualization Enables Explainable Validation of Metal-Organic Frameworks by Large Language Models

Interpretable Inverse Design of Metal-Organic Frameworks with Large Language Model Agents

MOF-VERIFY: A Failure-Aware Agentic Harness for MOF Hypothesis Verification