NAVER LABS Europe seminars are open to the public. Please register to participate.
Date: 13th February 2026, 11:00 am (CET)
About the speaker: Silvia Rossi is a full professor of Computer Science at the Department of Electrical Engineering and Information Technologies, University of Naples Federico II. She serves as the scientific director of the PRISCA Lab (Projects of Intelligent Robotics and Advanced Cognitive Systems). Prof. Rossi holds an M.Sc. in Physics from the University of Naples Federico II (2001) and a Ph.D. in Information and Communication Technologies from the University of Trento (2006). She has played a key role in numerous EU and international research projects and is currently the principal investigator and coordinator of several major initiatives, including the HORIZON-MSCA-2023-DN SWEET (Social aWareness for sErvicE roboTs), and HORIZON-TMA-MSCA-DN TRAIL (TRAnsparent, InterpretabLe Robots). Prof. Rossi chaired the RO-MAN conferences in 2020 and 2022 and is an active member of program committees for leading conferences in human-robot interaction and artificial intelligence. Her research focuses on Socially Assistive Robotics, Human-Robot Interaction, Cognitive Architectures, and User Profiling and Adaptive Systems. Her work explores computational methods for designing autonomous agents that can adapt their behavior to effectively interact with and support users.
Abstract: For service robots to successfully transition from controlled labs to complex, human-populated environments, they must evolve from seeing people as mere dynamic obstacles to recognizing them as social partners. Even in non-interactive tasks, effective and socially acceptable deployment requires perceiving and responding to human states, intentions, and emotions, and integrating these cues into the robot’s decision-making processes. At the same time, their own actions must be designed to “speak” clearly: movements that are legible, predictable, and easy for people to interpret. This talk will explore these two complementary dimensions—human-aware perception and human-understandable action. We first address the physical dimension of interaction, where robot movements must be designed to be “legible” and predictable—essentially “speaking” intent through action. However, physical legibility alone is insufficient without a foundation of deep situational awareness. Central to this framework is the integration of Theory of Mind (ToM), allowing robots to “mentalize” a user’s cognitive and affective states. Drawing on real-world applications like the BRILLO bartender, this talk provides a roadmap for developing robots capable of long-term, personalized, and socially aware operations.