!! Download poster and program (PDF): here 

Large Language Models (LLM)

May 20, 2025

Training morning

8.45am to 9am: Welcome greetings

9am to 12.30pm (including a break) : Use of LLMs

Gaël Lejeune is an Associate Professor in Computer Science at the Faculty of Arts of Sorbonne University, specializing in Natural Language Processing (NLP) and Image Analysis. He holds a PhD in Computer Science from the University of Caen on multilingual epidemiological monitoring (2013) and a habilitation to supervise research from Sorbonne University on the impact of data variation on the application of NLP methods (2023).

His research focuses on the robustness of NLP systems in various contexts of variation, whether linguistic (multilingualism, text genres, diachrony) or qualitative (noisy textual data obtained through web scraping, OCR, or ASR).

12.30pm to 2pm: Lunch

Afternoon of presentations

2pm to 3pm : Key-note speaker Christophe Benavent (Professor at Paris Dauphine - PSL)

AI, LLM and social sciences: a methodological perspective.

Starting in 2017, with the advent of Transformers, the field of Natural Language Processing (NLP) has rapidly advanced, leading to the emergence of very large generative models. This approach has extended beyond text processing and now enables multimodal document processing, opening new perspectives for corpus analysis in social sciences across four methodological domains: annotating documents for quantification and integration into econometric procedures, transforming documents for secondary processing, interacting with documents for interactive analysis, and finally, experimenting in silico. The main challenges revolve around the reliability of AI for these tasks, as well as the development of protocols to ensure the validity and reproducibility of measurements.

3pm to 3:40pm  :  Maria Manera (University of Turin, Bureau of Reasearch on Innovation Complexity and Knowledge  (BRICK)), Francesco Quatraro (University of Turin, Bureau of Reasearch on Innovation Complexity and Knowledge  (BRICK))

Mapping European Circular Economy Patents Using Deep Learning Models

3:40pm to 4pm  :  Pause

4pm to 4:40pm  :  Josep Domenech (Universitat Politècnica de València  (UPV)), Pablo De Pedraza (Universitat Politècnica de València  (UPV))

Using AI for Evaluating Public Policy Communication: Insights from the Comunitat Valenciana Case Study.

4:40m to 5:20pm  :  François Lafont (Centre d'Economie et de Sociologie Rurales Appliquées à l'Agriculture et aux Espaces Ruraux  (CESAER)), Eulalie Saïsset (Centre de Recherche sur les Inégalités Sociales  (CRIS), Sciences Po Paris)

Contested Ground: Exploring Land Use Conflicts in France.

5:20m to 6:00pm  : Del Puente Francesco (Parthenope, Italy), Yanhong Jin (Rutgers University, US), Alessandro Sapio (Parthenope, Italy), Carl Pray (Rutgers University, US)

 Assessing the impact of a European Union's policy on Agricultural Innovation in Italy.

Join us by visioconference for free (only on afternoon) !