The field of materials discovery is rapidly advancing with the integration of large language models (LLMs). Recent developments have shown that LLMs can be used to generate novel materials, optimize material properties, and accelerate the discovery process. The use of multimodal approaches, which combine LLMs with other techniques such as evolutionary computation and diffusion models, has been particularly effective in advancing the field. These approaches have been shown to outperform traditional methods and have the potential to revolutionize the field of materials discovery. Notable papers in this area include L2M3OF, which introduces a multimodal LLM for metal-organic frameworks, and REvolution, which combines LLMs with evolutionary computation for RTL generation and optimization. Additionally, papers such as LacMaterial and LLEMA have demonstrated the potential of LLMs for analogical reasoning and guided evolutionary search in materials discovery. Overall, the field of materials discovery is moving towards the development of more generalizable and powerful AI frameworks, such as GATE, which can be applied to a wide range of materials discovery tasks.
Advances in Materials Discovery with Large Language Models
Sources
L^2M^3OF: A Large Language Multimodal Model for Metal-Organic Frameworks
REvolution: An Evolutionary Framework for RTL Generation driven by Large Language Models
LacMaterial: Large Language Models as Analogical Chemists for Materials Discovery
Accelerating Materials Design via LLM-Guided Evolutionary Search
LLM Meets Diffusion: A Hybrid Framework for Crystal Material Generation
Towards a Generalizable AI for Materials Discovery: Validation through Immersion Coolant Screening
SHA-256 Infused Embedding-Driven Generative Modeling of High-Energy Molecules in Low-Data Regimes