The appendices list a large number of traits, and the species description paragraphs are quite lengthy.
As we want to test data extraction using LLMs, we might want to work on smaller subsets of data. It could be useful to assign the traits to high level categories (like stem, inflorescence, seed etc).
The appendices list a large number of traits, and the species description paragraphs are quite lengthy. As we want to test data extraction using LLMs, we might want to work on smaller subsets of data. It could be useful to assign the traits to high level categories (like stem, inflorescence, seed etc).