Extracting structured data from organic synthesis procedures using a fine-tuned large language model
Qianxiang Ai, Fanwang Meng, Jiale Shi, Brenden G. Pelkie, Connor W. Coley
Abstract
, 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.
Topics & Concepts
Schema (genetic algorithms)Computer scienceNatural language processingOpen sourceArtificial intelligenceInformation retrievalProgramming languageDatabaseSoftwareMachine Learning in Materials ScienceTopic ModelingBiomedical Text Mining and Ontologies