This research demonstrates how Large Language Models can address critical data limitations in materials science, specifically for graphene synthesis optimization.

• Tackles the challenge of scarce experimental data in materials science by leveraging LLMs
• Enables efficient extraction and structuring of synthesis data from scientific literature
• Combines discrete and continuous features effectively for improved machine learning models
• Creates a pathway for accelerating materials discovery and optimization despite limited datasets

This approach significantly reduces the engineering bottlenecks caused by costly and time-consuming experiments, potentially accelerating innovation cycles in advanced materials development.

Leveraging Large Language Models to Address Data Scarcity in Machine Learning: Applications in Graphene Synthesis