AI Seminar Cycle: Inside Multimodal Foundation Models – Representations as a Lever for Control, Safety and Grounding
AI Seminar Cycle: Inside Multimodal Foundation Models – Representations as a Lever for Control, Safety and Grounding
October 7 @ 11:00 AM - 12:00 PM
Join us for the opening session of the 2026–2027 Hi! PARIS × ELLIS AI Seminar Cycle, featuring Lorenzo Baraldi (University of Modena and Reggio Emilia).
In this talk, titled “Inside Multimodal Foundation Models: Representations as a Lever for Control, Safety, and Grounding,” Lorenzo will explore how internal representation spaces in multimodal models can enhance control, safety, and grounding in external knowledge, drawing on recent research from AImageLab and ELLIS Unit Modena.
