Harnessing Language Models for Experimental Materials Science

In this presentation we explore the use of large language models (LLM) and natural language processing (NLP) technique as a guide for experimentation in materials science. A specific focus of our discussion is the exploration of LLM and NLP methods to bridge disparate experimental information across multimodal experiments; especially to establish structure-processing-property relationships. We also discuss the application of LLM and NLP methods as a means to accelerate the discovery in identifying links and/or gaps in knowledge in experimental studies.