Tracing the Genealogies of Darwinian Ideas with LLM embeddings
Li, Lucian
UIUC School of Information Science, United States of America
I propose a method to detect similar ideas across a large text corpus using embeddings to capture semantic meaning and argumentative structure. I create a corpus of 100,000 nonfiction academic books from the 19th century and trace genealogies of ideas found in Darwin's publications.
Commoning Biomedicine: An application of LLM technology to support historians of science
Belouin, Pascal; El-Hajj, Hassan; Freeborn, Alfred
Max Planck Institute for the History of Science, Germany
The Commoning Biomedicine project aims to improve accessibility to a number of heterogenous oral history collections in biomedicine.
Integrating Large Language Model technology to our platform allowed us to improve metadata extraction and transcript summarization processes. Our talk will discuss the benefits of this approach and the challenges we faced.
La retorica dei Large Language Models: questioni etiche
Raffini, Daniel
Sapienza Università di Roma, Italy
L’intervento analizza in prospettiva etica la reotrica dei testi prodotti tramite LLMs. Si prenderanno in considerazione due aspetti: la simulazione dell'agency umana, che stimola una reazione empatica ingannevole, e l’utilizzo discriminatorio della lingua. L’analisi retorica mira ad affiancare l’etica dell’AI, fornendo uno strumento per intervenire sui sistemi.
What’s in an entity? Exploring Nested Named Entity Recognition in the Historical Land Register of Basel (1400-1700).
Prada Ziegler, Ismail (1,2)
1: University of Bern, Switzerland; 2: University of Basel, Switzerland
HTML XMLThe Historical Land Register of Basel contains information about property transactions from late medieval to early modern times. In this submission I report my findings applying machine learning techniques to extract nested named entities from this pre-modern German-language corpus.
Leveraging Large Language Models to Generate a Knowledge Graph for Italian Literary Texts
Santini, Cristian; Marozzi, Gioele; Frontoni, Emanuele; Melosi, Laura
University of Macerata, Italy
HTML XMLThis paper outlines a methodology for Knowledge Extraction from historical literary texts in Italian using a combination of Large Language Models and fine-tuned models for Relation Extraction. The research aims to offer a novel way to extract and represent entities and relations from literary manuscripts.