Fine-Tuned Language Models Generate Stable Inorganic Materials as Text
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | , , , , , |
|---|---|
| Format: | Preprint |
| Publié: |
2024
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866911073286750208 |
|---|---|
| author | Gruver, Nate Sriram, Anuroop Madotto, Andrea Wilson, Andrew Gordon Zitnick, C. Lawrence Ulissi, Zachary |
| author_facet | Gruver, Nate Sriram, Anuroop Madotto, Andrea Wilson, Andrew Gordon Zitnick, C. Lawrence Ulissi, Zachary |
| contents | 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. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2402_04379 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Fine-Tuned Language Models Generate Stable Inorganic Materials as Text Gruver, Nate Sriram, Anuroop Madotto, Andrea Wilson, Andrew Gordon Zitnick, C. Lawrence Ulissi, Zachary Machine Learning Materials Science 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. |
| title | Fine-Tuned Language Models Generate Stable Inorganic Materials as Text |
| topic | Machine Learning Materials Science |
| url | https://arxiv.org/abs/2402.04379 |