[20a-A21-1~10] 23.1 Joint Session N "Informatics"

11:15 AM - 11:30 AM JST(2:15 AM - 2:30 AM UTC)

[20a-A21-9] Materials Development with LLM (I): Development Scope through Model Comparison

〇Masashi Ishii¹, Yasufumi Watanabe¹, Hiroyuki Oka¹, Hiroyuki Shindo² (1.NIMS, 2.MatBrain)

Keywords:

Large language model,Materials Development,Knowledge extraction

LLM is a language model trained with a neural network on a large amount of data. In addition to the previous automatic data extraction using natural language processing, we have now started automatic knowledge extraction. In other words, we aim to formulate guidelines for material development, extract issues and solutions, identify examples and optimal solutions, and construct and infer material models by LLM, and investigate issues, develop methods, and verify them. The scope of this effort is presented here through model comparison.