Prof. Alaa Maalouf
Neubauer Assistant Professor of Computer Science, University of Haifa
IAS SEMINAR #16/2026
Turin / OGR / Mezzanino / September 29, 2026 / 04:00 – 06:00 pm CEST
Pixels as Concepts: Toward Open-Set and Generalizable Multimodal Physical AI
EVENT&WEBINAR
Physical AI must understand more than a predefined vocabulary of objects—it must interpret an open-ended world through concepts, language, geometry, and interaction, while generalizing to environments and conditions far beyond those seen during training. This talk presents a vision in which every pixel carries semantic meaning. By grounding multimodal foundation-model representations in images, video, and 3D space, robots can connect what they see with what humans describe—and use that connection to act in unfamiliar environments.
I will show how shared conceptual representations can enable open-set object following, generalizable autonomous driving, text-queryable 3D mapping, and language-guided navigation. Rather than learning every task and environment separately, these systems reuse a common representation across perception, mapping, and control.
Finally, I will show how the same idea extends beyond robotics to scientific discovery. In Project CETI, this idea of treating pixels as concepts served as a building block to transform rare drone footage of a sperm-whale birth into quantitative behavioral evidence—revealing coordinated caregiving across kinship lines and demonstrating how AI can convert complex natural observations into testable biological insight.
ALAA MAALOUF
Dr. Alaa Maalouf is the Neubauer Assistant Professor of Computer Science at the University of Haifa, where he directs the GOATS Lab—Generalizable & Optimized AI for Technology and Science. He is also a Research Affiliate at MIT CSAIL, an Associate at Harvard SEAS, and a Principal Investigator in Project CETI. Before joining the University of Haifa, he was a postdoctoral researcher at MIT CSAIL, hosted by Prof. Daniela Rus. His research spans efficient and reliable AI, physical AI, and AI-driven scientific discovery.
His work has received distinctions including a NeurIPS 2025 Spotlight, an ICRA 2026 Oral presentation, a NeurIPS 2019 Outstanding Paper Honorable Mention, and a CVPR 2021 Outstanding Reviewer Award. He has received the CHE MAOF Award for outstanding young faculty, the Neubauer Family Foundation Fellowship for young faculty, the University of Haifa’s Faculty of Computer Science and Information Systems Award for Research Excellence, and the Rector’s Prize for Outstanding Young Researcher. His research has also received broad international coverage, including in The New York Times, National Geographic, Nature,Reuters, CNN, AP, NPR, and Scientific American. He is also a TEDx speaker.
In addition to academia, Dr. Maalouf spent five years in industry, serving as a Lead ML Researcher at Data Heroes Ltd. and as a chip-design engineer at Mellanox Technologies (now part of NVIDIA). During his PhD and postdoctoral work, he also collaborated extensively with industry on real-world problems, including projects with Samsung Research, the Toyota Research Institute, and several startups.
Register for In-Person Attendance
Joining us in person offers the opportunity for direct networking and interaction with both speakers and participants.
Please complete the form below to reserve your seat.
By clicking, you will be automatically registered for the in-person event.
Note: If you see a message saying you’re already subscribed, please ignore it — your in-person registration will still be confirmed.
Please note that, by attending, you implicitly authorize AI4I to film and photograph the audience during the seminar. By submitting your registration, you acknowledge and accept the AI4I Data Privacy Policy. Seats are limited: if you are unable to attend, we kindly ask you to inform us as soon as possible at ias@ai4i.it
LINK ZOOM
