Foundation Models and LLMs for Materials Science: A Review of Prompting, Fine-Tuning, and What Does Not Transfer

iamrp.net-cOaX-1790645728-w258707955 1790645728 Institute for Advanced Materials Research Press info@iamrp.net Institute for Advanced Materials Research Press Journal of Computational and Data-Driven Materials Engineering J. Comput. Data-Driven Mater. Eng. 3149-9368 2026 5 1 Foundation Models and LLMs for Materials Science: A Review of Prompting, Fine-Tuning, and What Does Not Transfer Gabriel Costa Rafael Mendes Bruno Teixeira Lucas Ribeiro 2026 10.68159/w258707955 https://iamrp.net/pub/journal/2/article/w258707955 Pyzer-Knapp EO, Manica M, Staar P, Morin L, Ruch P, Laino T, et al. Foundation models for materials discovery-current state and future directions. NPJ Comput Mater. 2025;11(1):61. 10.1038/s41524-025-01538-0 Batatia I, Benner P, Chiang Y, Elena AM, Kovács DP, Riebesell J, et al. A foundation model for atomistic materials chemistry. J Chem Phys. 2025;163(18):184110. 10.1063/5.0297006 Soares EA, Vital Brazil EA, Shirasuna V, Zubarev D, Cerqueira RFG, Schmidt K. A Mamba-based foundation model for materials. NPJ Artif Intell. 2025;1(1):8. 10.1038/s44387-025-00009-7 Zhang L, Liu Z, Ni B, Wang Q. Large language models (LLMs) for materials design. Adv Funct Mater. 2026;36(30):e25897. 10.1002/adfm.202525897 Niyongabo Rubungo A, Li K, Hattrick-Simpers J, Dieng AB. LLM4Mat-bench: Benchmarking large language models for materials property prediction. Mach Learn Sci Technol. 2025;6(2):020501. 10.1088/2632-2153/add3bb Niyongabo Rubungo A, Arnold CB, Rand BP, Dieng AB. LLM-Prop: Predicting the properties of crystalline materials using large language models. NPJ Comput Mater. 2025;11(1):186. 10.1038/s41524-025-01536-2 Wang H, Li K, Ramsay S, Fehlis Y, Kim E, Hattrick-Simpers J. Evaluating the performance and robustness of LLMs in materials science Q&A and property predictions. Digit Discov. 2025;4(6):1612-24. 10.1039/D5DD00090D Jablonka KM, Ai Q, Al-Feghali A, Badhwar S, Bocarsly JD, Bran AM, et al. 14 examples of how LLMs can transform materials science and chemistry: A reflection on a large language model hackathon. Digit Discov. 2023;2(5):1233-50. 10.1039/D3DD00113J Gupta S, Mahmood A, Shukla S, Ramprasad R. Benchmarking large language models for polymer property predictions. Macromol Rapid Commun. 2025:e00388. 10.1002/marc.202500388 Yoshitake M, Nagata T. A method for LLM-based construction of a materials property knowledge graph: A case study. Appl Sci. 2025;15(19):10511. 10.3390/app151910511 Liu S, Wen T, Pattamatta ASLS, Srolovitz DJ. A prompt-engineered large language model, deep learning workflow for materials classification. Mater Today. 2024;80:240-9. 10.1016/j.mattod.2024.08.028 Zhou J, Xiao B, Liu Q, Liu L, Zhang L. MatPC: Prompting large language model, crystal structure prediction, and first-principles for semantic-driven material design. ACS Appl Mater Interfaces. 2025;17(31):44528-40. 10.1021/acsami.5c08809 Radova M, Stark WG, Allen CS, Maurer RJ, Bartók AP. Fine-tuning foundation models of materials interatomic potentials with frozen transfer learning. NPJ Comput Mater. 2025;11(1):237. 10.1038/s41524-025-01727-x Kong L, Shoghi N, Hu G, Li P, Fung V. MatterTune: An integrated, user-friendly platform for fine-tuning atomistic foundation models to accelerate materials simulation and discovery. Digit Discov. 2025;4(8):2253-62. 10.1039/D5DD00154D Hwang J, Lee T, Lee Y, Yoo SH. Fine-tuning bulk-oriented universal interatomic potentials for surfaces: Accuracy, efficiency, and forgetting control. arXiv [Preprint]. 2025. 10.48550/arXiv.2509.25807 Feng M, Zhao C, Day GM, Evangelopoulos X, Cooper AI. A universal foundation model for transfer learning in molecular crystals. Chem Sci. 2025;16(28):12844-59. 10.1039/D5SC00677E Novelli P, Bonati L, Buigues PJ, Meanti G, Rosasco L, Parrinello M, et al. Fine-tuning foundation models for molecular dynamics: A data-efficient approach with random features. In: Proceedings of the Advances in Neural Information Processing Systems; 2024. Albakri B, Kister A, Benner P. Exploring transfer learning for materials property prediction. In: AI for Accelerated Materials Design Workshop, ICLR 2026; 2026. Cho Y, Yi S, Yang W, Kang S, Son YW, Choo J, et al. Robust and interpretable adaptation of equivariant materials foundation models via sparsity-promoting fine-tuning. arXiv [Preprint]. 2026. 10.48550/arXiv.2606.18691 Menon SS, Mondal T, Brahmachary S, Panda A, Joshi SM, Kalyanaraman K, et al. On scientific foundation models: Rigorous definitions, key applications, and a comprehensive survey. Neural Netw. 2026;198:108567. 10.1016/j.neunet.2026.108567 Durmaz AR, Lamb JD, Echlin MP, Pollock TM. Foundation models for multimodal image data fusion in materials science. Front Mater. 2026;13:1815017. 10.3389/fmats.2026.1815017 Chaudhari A, Ock J, Barati Farimani A. Modular large language model agents for multi-task computational materials science. Commun Mater. 2026;7(1):131. 10.1038/s43246-025-00994-x Yuan ECY, Liu Y, Chen J, Zhong P, Raja S, Kreiman T, et al. Foundation models for atomistic simulation of chemistry and materials. Nat Rev Chem. 2026;10(3):212-30. 10.1038/s41570-025-00793-5 Zhang T, Ladhak F, Durmus E, Liang P, McKeown K, Hashimoto TB. Benchmarking large language models for news summarization. Trans Assoc Comput Linguist. 2024;12:39-57. 10.1162/tacl_a_00632 Xie T, Wan Y, Huang W, Zhou Y, Liu Y, Linghu Q, et al. Large language models as master key: Unlocking the secrets of materials science with GPT. arXiv [Preprint]. 2023. 10.48550/arXiv.2304.02213 Mittelstadt B, Wachter S, Russell C. To protect science, we must use LLMs as zero-shot translators. Nat Hum Behav. 2023;7(11):1830-2. 10.1038/s41562-023-01744-0 Brown TB, Mann B, Ryder N, Subbiah M, Kaplan JD, Dhariwal P, et al. Language models are few-shot learners. Adv Neural Inf Process Syst. 2020;33:1877-901. Axelrod S, Schwalbe-Koda D, Mohapatra S, Damewood J, Greenman KP, Gómez-Bombarelli R. Learning matter: Materials design with machine learning and atomistic simulations. Acc Mater Res. 2022;3(3):343-57. 10.1021/accountsmr.1c00238 Yang J, Yin Z, Ao L, Li S. MACE foundation models for lattice dynamics: A benchmark study on double halide perovskites. Phys Chem Chem Phys. 2026;28(7):4459-69. 10.1039/D5CP04693A Ma Y, Ren Q, Hettinga K, Fogliano V. Leveraging foundation models and transfer learning for peptide transport prediction, molecular taste classification, and visual texture analysis. Innov Food Sci Emerg Technol. 2025;105:104247. 10.1016/j.ifset.2025.104247 Zhang X, Zhang P, Yuan J, Li L. Zero-shot learning for materials science texts: Leveraging duck typing principles. In: Proceedings of the AAAI Conference on Artificial Intelligence. 2025;39(1):1129-37. 10.1609/aaai.v39i1.32100 Liu H, Yin H, Luo Z, Wang X. Integrating chemistry knowledge in large language models via prompt engineering. Synth Syst Biotechnol. 2025;10(1):23-38. 10.1016/j.synbio.2024.07.004 Takeda S, Kishimoto A, Hamada L, Nakano D, Smith JR. Foundation model for material science. Proc AAAI Conf Artif Intell. 2023;37(13):15376-83. 10.1609/aaai.v37i13.26793 Gao YC, Chen X, Yuan YH, Chen YP, Niu YL, Yao N, et al. Accelerating battery innovation: AI-powered molecular discovery. Chem Soc Rev. 2025;54(21):9630-84. 10.1039/D5CS00053J Takeda S, Priyadarsini I, Kishimoto A, Shinohara H, Hamada L, Masataka H, et al. Multi-modal foundation model for material design. In: AI for Accelerated Materials Design Workshop, NeurIPS 2023; 2023.