Davide Buscaldi is associate professor at Université Sorbonne Paris Nord and “chargé d’enseignement” (part-time) at DIX, Ecole Polytechnique. He has acquired an important experience with text mining for the construction of Knowledge Graphs for the construction of the Computer Science Knowledge Graph (CS-KG) , a joint project with University of Cagliari (Italy) and The Open University (UK). For this project he supervised the construction of a pipeline that extracts triples from scientific texts and verify their consistency with the help of a neural network.

He is the author of more than 120 works related to NLP and text mining, published in peer-reviewed international and national venues. He co-directed 4 Ph.D. at LIPN and 1 Ph.D. at LIX and he is currently co-directing 2 more Ph.D. at LIX on Large Language Models and a Ph.D. at LIPN on Graph Neural Networks.

Message from Prof. Buscaldi:

In this talk, I will show some preliminary results on the surge of the use of chatGPT by students in French Computer Science BUT at Villetaneuse. I will discuss the capabilities and the reliability of AI detection tools, then I will highlight how these tools can be used to estimate the use of AI and chatGPT to write reports, and finally I will give some insights regarding which parts of these reports tend to be generated automatically.