Large Language Model-augmented taxonomy for >4500 palaeopalynology genera

Le résumé fourni par la source

Large Language Models (LLMs), being text-based, are ideal types of artificial intelligence to analyze the complexities of palaeontological taxonomy because paleontology depends on published textual descriptions as the primary, authoritative record of a taxon. This paper describes: (1) the preparation of palynological taxonomic text contained within the > 4500 genera of the Jansonius and Hills palaeopalynological catalog (JHC) for an LLM-augmented taxonomy system (LATS) that aids taxonomic determination: (2) the efficiency and accuracy of the LATS: and (3) examples of possible further uses of the LATS beyond aids to identification. Principles of development include: (1) ‘inclusion’, that is, making sure that the LATS as far as possible includes rather than excludes candidate genera; (2) the principle of ‘assistance’ rather than supplantation so that the LATS is intended as an aid to taxonomy, not a replacement for a human taxonomist; and (3) the principle of ‘non-intervention’ whereby no alterations to original authoritative genus descriptions or diagnoses are applied. The LATS functions through Retrieval Augmented Generation and returns candidate genera with statistical measures of match against the prompt(s). The LATS produces generally good results but there are two types of limitations or shortcomings: those that emanate from the JHC (and palaeopalynological taxonomy itself), and those that emanate from the working of the LATS. As well as providing a LATS to aid a palynologist through the stages of identification, the information and ‘understanding’ that the underlying system has of a large area of palaeopalynological taxonomy means that the system could be put to more general uses: for example in the identification of genus names that are likely synonyms, and investigation of the distribution of genera (or taxa) in ‘morphological space’.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.