LATHE: LAnguage-driven Toolkit for Hypothesis-based crystal Editing
Abstract
Manipulating crystal structures towards targeted geometric properties or symmetry constraints is a longstanding challenge. Existing computational frameworks fall short: LLMs have poor spatial awareness and operate at coarse granularity in the token space, while diffusion- and gradient-based structure generation approaches generally produce complete structures \textit{de novo} rather allowing for fine-grained editing of a given input under arbitrary constraints. We present LATHE, a geometric toolkit that closes this gap by expressing seven classes of crystallographic properties --- bond length, bond angle, dihedral angle, coordination environment, lattice parameters, cell volume, and space group, as differentiable objectives to enable direct geometric editing of crystal structures. To demonstrate its usage in materials design, we further expose LATHE through a Model Context Protocol server and embed it in a closed-loop multi-agent system which translates hypotheses intogeometric modifications in natural language towards a given design objective. Across single-property benchmarks, LATHE attains near-perfect constraint satisfaction on all seven property types while keeping optimized structures close to local energy minima. The LATHE-equipped agent translates over 87.5\% of natural-language prompts into valid executable configurations and faithfully completes them. In a band-gap inverse-design case study, the multi-agent loop reaches the target tolerance window in nine of ten independent runs at a typical cost of thirteen hypothesis-evaluation cycles. By bridging natural-language hypothesis generation and physically grounded gradient-based structural editing, this work establishes a paradigm for interpretable, closed-loop computational materials discovery that is immediately applicable to a broad class of functional material design tasks.
Related
- BootLoops: an LLM-driven toolkit for exact quantitative science
- OpenULF: A Modular Toolkit Enables Data-Driven Ultra-Low-Field MRI with both Classical and Quantum Sensors
- Separating perception from reasoning in vision-language models: a model-free render ceiling for crystal structures
- Oscillator-driven broadband femtosecond transient microspectroscopy with supercontinuum probes generated in a photonic crystal fiber
- Crystal-structure design by agentic AI in a language of motifs
BibTeX
@misc{zheng2026lathe,
title = {{LATHE: LAnguage-driven Toolkit for Hypothesis-based crystal Editing}},
author = {Qianyu Zheng and Shuyi Jia and Victor Fung},
howpublished = {arXiv},
year = {2026},
month = oct,
doi = {10.48550/arxiv.2610.02671},
eprint = {2610.02671},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2610.02671},
note = {CC BY 4.0},
}
RIS
TY - UNPB
TI - LATHE: LAnguage-driven Toolkit for Hypothesis-based crystal Editing
AU - Qianyu Zheng
AU - Shuyi Jia
AU - Victor Fung
PY - 2026
DA - 2026/10/02
PB - arXiv
DO - 10.48550/arxiv.2610.02671
UR - https://arxiv.org/abs/2610.02671
AB - Manipulating crystal structures towards targeted geometric properties or symmetry constraints is a longstanding challenge. Existing computational frameworks fall short: LLMs have poor spatial awareness and operate at coarse granularity in the token space, while diffusion- and gradient-based structure generation approaches generally produce complete structures \textit{de novo} rather allowing for fine-grained editing of a given input under arbitrary constraints. We present LATHE, a geometric toolkit that closes this gap by expressing seven classes of crystallographic properties --- bond length, bond angle, dihedral angle, coordination environment, lattice parameters, cell volume, and space group, as differentiable objectives to enable direct geometric editing of crystal structures. To demonstrate its usage in materials design, we further expose LATHE through a Model Context Protocol server and embed it in a closed-loop multi-agent system which translates hypotheses intogeometric modifications in natural language towards a given design objective. Across single-property benchmarks, LATHE attains near-perfect constraint satisfaction on all seven property types while keeping optimized structures close to local energy minima. The LATHE-equipped agent translates over 87.5\% of natural-language prompts into valid executable configurations and faithfully completes them. In a band-gap inverse-design case study, the multi-agent loop reaches the target tolerance window in nine of ten independent runs at a typical cost of thirteen hypothesis-evaluation cycles. By bridging natural-language hypothesis generation and physically grounded gradient-based structural editing, this work establishes a paradigm for interpretable, closed-loop computational materials discovery that is immediately applicable to a broad class of functional material design tasks.
ER -