How Continual Learning and Generative AI
are transforming Robotics
Workshop @ I-RIM 3D 2025 Conference
October 17, 2025 - 3pm to 5.30pm CEST - Gazometro Ostiense, Rome, Italy
Abstract
Continual Learning and Generative AI are among the most transformative developments in modern machine learning. Continual Learning enables systems to learn from sequential data over time without catastrophic forgetting, while Generative AI technologies - such as large language and vision models or diffusion and generative models - are changing how robots perceive, reason, and act with minimal supervision.
This workshop aims to explore how these two paradigms can be effectively integrated into Robotics to support long-term autonomy, adaptability, and scalable intelligence. At the same time, we will address the challenges that arise when deploying these techniques in embodied systems, including computational limitations, safety, and real-world generalization.
Topics include:
- Continual learning algorithms for robotics
- Generative models for perception, planning, and control
- Benchmarks and evaluation protocols
- Real-world applications in service, assistive, and industrial robotics
The workshop will feature invited talks, a poster and pitch session, and an open discussion to foster collaboration between the AI and robotics communities.
Organizers
|
|
|
PostDoctoral Researcher BRAIR Lab @ The BioRobotics Institute, Scuola Superiore Sant'Anna, Italy | Associate Professor Department of AI, Data and Decision Sciences, LUISS Guido Carlo University, Italy | Associate Professor BRAIR Lab @ The BioRobotics Institute, Scuola Superiore Sant'Anna, Italy |
Invited speakers (confirmed)
Tatiana Tommasi Associate Professor 
ELLIS Unit Department of Control and Computer Engineering, Politecnico di Torino, Italy | Talk: Generative AI for Robot Grasping Bio and Research Activities Tatiana Tommasi received the PhD degree from EPFL Lausanne (CH), in 2013. She is associate professor with the Department of Control and Computer Engineering, Polytechnic of Turin (IT), affiliated researcher with the Italian Institute of Technology, and director with the ELLIS Unit in Turin.She has published more than 50 papers at top conferences and journals in machine learning and computer vision. She has a strong record in theoretically grounded algorithms for automatic learning from images with robotics, medical, and human-machine interaction applications. She pioneered the area of transfer learning in computer vision and has extensive experience in domain adaptation, generalization, multimodal, and open-set learning. |
Loris Roveda Associate Professor 
Mechanical Department, Politecnico di Milano, Italy & Scuola Universitaria Professionale della Svizzera Italiana, Switzerland | Talk: Unleashing Autonomy for Real-World Robots Bio and Research Activities Loris Roveda received his PhD from Politecnico di Milano in 2015. He has been a researcher at STIIMA-CNR from 2015 to 2019 and a senior researcher at IDSIA USI-SUPSI from 2019. He is an Associate Professor at Politecnico di Milano, Mechanical Department, working in robotics and machine learning. His main research interests find applications in exoskeleton design and control, human-robot collaboration, autonomous robotics, reinforcement learning, generative AI, and industrial process modeling, optimization, and control. He is a co-author of international scientific papers published in conferences and journals. He is involved in national and international funded projects and collaborations. |
Matteo Saveriano Associate Professor 
Department of Industrial Engineering, University of Trento, Italy | Talk: Towards Safe Continual Learning from Demonstration Bio and Research Activities Matteo Saveriano worked as a research assistant at Technical University of Munich where he got a Ph.D. in 2017. He later worked as a post-doctoral researcher at the German Aerospace Center (DLR), and at the Department of Computer Science and at the Digital Science Center of the University of Innsbruck. Until 2021, he was an assistant professor at the Department of Computer Science and at the Digital Science Center of the University of Innsbruck. He is currently an associate professor at the Department of Industrial Engineering of the University of Trento. His research attempts to integrate cognitive robots into smart factories and social environments through the embodiment of AI solutions, inspired by the human behavior, into robotic devices. From the economical perspective, cognitive robots have the potential to significantly increase the productivity and the flexibility of modern enterprises. From the social point of view, robots with embodied AI capabilities will improve the working conditions and will guarantee seamless care and assistance to people in need. |
Vincenzo Lomonaco Associate Professor
Department of AI, Data and Decision Sciences, LUISS Guido Carlo University, Italy | Talk: The Future of Continual Learning in the Era of Foundation Models: Three Key Directions Bio and Research Activities Vincenzo Lomonaco is an Associate Professor at the LUISS Guido Carlo University in Rome, Italy. He was formely a Researcher at the University of Pisa, Italy where he taught the Artificial Intelligence course. Currently, he also serve as Co-Founding President at ContinualAI, a non-profit research organization and the largest open community on Continual Learning for AI, Co-founding Board Member at AI for People, and as Co-Founder of ContinualIST: a University of Pisa spin-off. In Pisa, he worked within the Pervasive AI Lab as Task Leader of two main European projects and as Principal Investigator of several industrial research contracts with companies such as Meta, Intel, Leonardo s.p.a. and SeaVision s.r.l. He is among the 350 Italian researchers of Future Artificial Intelligence Research (FAIR), the largest Italian initiative on AI with a budget of over 110 millions and proud member of ELLIS and CLAIR. His main research interest and passion is about Continual Learning in all its facets. In particular, he loves to study Continual Learning under four main lights: Deep Learning, Distributed Learning and Practical Applications, all within a AI Sustainability developmental framework |
Egidio Falotico Associate Professor
BRAIR Lab The BioRobotics Institute, Scuola Superiore Sant'Anna, Italy | Talk: Continual Learning and Generative AI in Soft Robotics Bio and Research Activities Egidio Falotico received the M.S. degree in computer science from the University of Pisa, Italy in 2008 and the Ph.D. degree in biorobotics from Scuola Superiore Sant’Anna (SSSA), Italy in 2013, and the Ph.D. degree in cognitive science fromthe University Pierre et Marie Curie, France in March 2013. He is currently an Associate Professor with The BioRobotics Institute, SSSA. He is the author or coauthor of more than 100 international peer-reviewed papers and he regularly serves as a reviewer for more than 10 international ISI journals. He has been involved in some EU-fundedprojects (PROBOSCIS, Growbot, I-SUPPORT, SWARMs, SMARTE, RoboSoM, RobotCub), focusing on the development of brain-inspired algorithms for robot control. His research interests focus on neurorobotics - the implementation of brain models from neuroscience in robots - and the deployment of learning-based control algorithms on rigid and soft robots, for collaborative and contact-rich tasks. |
Program (3pm to 5.30pm CEST)
Time | Topic | Speaker |
| 3pm - 3.10pm | Welcome and Introduction | Organizers |
| 3.10pm - 3.20pm | Talk: "The Future of Continual Learning in the Era of Foundation Models: Three Key Directions" | Vincenzo Lomonaco |
| 3.20pm - 3.30pm | Talk: "Towards Safe Continual Learning from Demonstration" | Matteo Saveriano |
| 3.30pm - 3.40pm | Talk: "Continual Learning and Generative AI in Soft Robotics" | Egidio Falotico |
| 3.40pm - 3.50pm | Talk: "Generative AI for Robot Grasping" | Tatiana Tommasi |
| 3.50pm - 4.00pm | Talk: "Unleashing Autonomy for Real-World Robots" | Loris Roveda |
| 4.00pm - 4.20pm | Discussion with experts | All, Enrico Donato (chair) |
| 4.20pm - 4.30pm | Pitch of selected contributions | |
| 4.30pm - 5.30pm | Poster session |
Call for Contributions
Link for submission - Deadline: September 23, 2025
We invite young researchers and students to submit an extended abstract including contributions, but not limited to, on (i) Continual Learning for Robotics, (ii) Generative AI for perception, planning, and control, and (iii) advanced learning strategies and benchmarks.
Submission guidelines are the same as those outlined in the regular call for contributions. When uploading your work, please select the presentation option “How Continual Learning and Generative AI are Transforming Robotics”.
Authors of selected abstracts will be invited to present their work with a pitch talk and a poster (A0 size, portrait) during the poster session.
If you have additional questions, please contact: enrico.donato@santannapisa.it