Extracting structured data from organic synthesis procedures using a fine-tuned large language model

作者：Qianxiang Ai, Fanwang Meng, Jiale Shi, Brenden G. Pelkie, Connor W. Coley · 发表于：Digital Discovery · 年份：2024 · DOI：10.1039/d4dd00091a · 被引用次数：33 · 研究领域：Machine Learning in Materials Science、Topic Modeling、Biomedical Text Mining and Ontologies

, compound identifiers, mass quantities), with the ability to recognize compound-referencing tokens and to infer reaction roles. We investigate its failure modes and evaluate performance on specific subtasks such as reaction role classification.