Sara Candussio

PhD student in Natural language processing University of Trieste

Sara Candussio is a PhD student in Natural Language Processing at the University of Trieste (AI Lab), supervised by Luca Bortolussi and by Gabriele Sarti, a postdoctoral researcher at Northeastern University. Her research focuses on reasoning in Large Language Models. During her PhD she spent a period as a visiting researcher at the Center for Language and Cognition (CLCG) of the University of Groningen. She holds a Bachelor’s degree in Statistics and Computer Science and a Master’s degree in Data Science and Scientific Computing from the University of Trieste and SISSA, both with highest honours. Her Master’s thesis “A Dialectic Pipeline for Improving LLM Robustness” received an honourable mention at the “Emanuele Pianta” Award (AILC, CLiC-it 2025). She worked as a Research Scientist at Rachael S.r.l. on applied Natural Language Processing and synthetic data generation. She is a teaching assistant for the Probabilistic Machine Learning course, Vice President of the AI Student Society (AI2S), and departmental representative for MIGe.

Sun. 27 September

at 15:00 - 16:15

15:00 - 16:15

Sunday 27 September

Area Talk 2
Piazza Unità d'Italia, 34121, Trieste, TS

How does an artificial intelligence take decisions? And does our brain really work so differently? Advanced algorithms now forecast the weather and analyse complex data, yet they often remain hard to interpret. Three PhD students guide the audience through the new intelligences, from AI to neuroscience, looking at more transparent models, explainable predictions and the workings of the nervous system. A tour of research supported by the European Social Fund Plus (ESF+) 2021–27, exploring how thought arises and what it means for everyday life.