REVIEW 8 cited by
Fine-Tuned Language Models Generate Stable Inorganic Materials as Text
Not yet reviewed by Pith; the record is open.
This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.
SPECIMEN: schema-true, not a live event
T0 review · schema-true
One-sentence machine reading of the paper's core claim.
pith:XXXXXXXX · record.json · timestamp
read the original abstract
We propose fine-tuning large language models for generation of stable materials. While unorthodox, fine-tuning large language models on text-encoded atomistic data is simple to implement yet reliable, with around 90% of sampled structures obeying physical constraints on atom positions and charges. Using energy above hull calculations from both learned ML potentials and gold-standard DFT calculations, we show that our strongest model (fine-tuned LLaMA-2 70B) can generate materials predicted to be metastable at about twice the rate (49% vs 28%) of CDVAE, a competing diffusion model. Because of text prompting's inherent flexibility, our models can simultaneously be used for unconditional generation of stable material, infilling of partial structures and text-conditional generation. Finally, we show that language models' ability to capture key symmetries of crystal structures improves with model scale, suggesting that the biases of pretrained LLMs are surprisingly well-suited for atomistic data.
Forward citations
Cited by 8 Pith papers
-
MBFormer: A General Transformer-based Learning Paradigm for Many-body Interactions in Real Materials
MBFormer maps DFT mean-field states to GW quasiparticle energies and BSE exciton properties, achieving 0.16 eV and 0.20 eV MAE on held-out 2D materials and extrapolating from coarse to fine k-grids.
-
AGAPI-Agents: An Open-Access Agentic AI Platform for Accelerated Materials Design on AtomGPT.org
An open-source agentic materials-design platform shows tool access can help or hurt accuracy depending on the property, but its headline memorization-resistant test results are not presented.
-
MiAD: Mirage Atom Diffusion for De Novo Crystal Generation
Mirage infusion lets crystal diffusion models vary atom counts during generation and raises the S.U.N. rate on MP-20 to 8.2%.
-
Teaching LLMs to Speak Spectroscopy
A LLaMA-3.1-8B model fine-tuned with LoRA on digit-serialized SDSS spectra predicts redshifts with MAE 0.043 and retains 85% of its astronomy QA performance.
-
Enhancing Materials Discovery with Valence Constrained Design in Generative Modeling
CrysVCD generates valence-balanced compositions with an elemental language model and then constructs their crystal structures with a diffusion model, reporting improved stability and functional property targeting.
-
VASP Plugins: Linking the Vienna ab-initio Simulation Package with Python
A C++/pybind11 shared-memory plugin layer exposes VASP SCF and ionic data as NumPy arrays so Python can modify structure, forces, local potential, and occupancies in place.
-
AtomBench: A Benchmarking Framework for Generative Crystal Reconstruction Models in Conventional Superconductors
On two superconductor datasets, CDVAE best reproduces lattice parameters while AtomGPT (full text) or MatterGen (abstract) best reproduces atomic coordinates, but the comparison is confounded by unequal input information.
-
A Survey of AI for Materials Science: Foundation Models, LLM Agents, Datasets, and Tools
This survey organizes foundation models, LLM agents, datasets, and tools in materials science into six task areas.
Discussion (0). Sign in to comment.