People
People

Person details

Paolo Riva is a Ph.D. student in Information Technology at Politecnico di Milano, conducting his doctoral studies at the Artificial Intelligence and Robotics Laboratory (AIRLab) under the supervision of Prof. Matteo Matteucci.
He received a B.Sc. (2023) from the University of Brescia and an M.Sc. (2026) from Politecnico di Milano in Computer Science and Engineering, where he developed his master's thesis work titled "Language-driven robot spatial understanding with long-horizon semantic graph memory". The project focused on spatial-semantic open-world perception and memory representation in real-time robot scenarios, introducing novel embodied agent architectures leveraging Vision-Language Models (VLMs) and Large Language Models (LLMs) in robot perception, reasoning, and human interaction.
His current research focuses on the intersection of Agentic AI and Robotics, integrating embodied agent architectures and foundation models in robotic applications for generalized reasoning at the edge. Specifically, his work centers on real-time VLM perception, Retrieval-Augmented Generation (RAG) in LLM-based robotic agents, and World Models for planning and outcome prediction.
His research vision is to contribute to a future where intelligent robots integrate seamlessly and safely into daily life, grounding long-horizon learning capabilities in deployable, real-world solutions to benefit society.