AIRoV Keynotes

At AIROV 2026, we are proud to feature three distinguished keynote speakers who bring together cutting-edge technical insights and legal perspectives on AI and robotics. Please find below more detailed info on the respective keynotes as well as the authors.

Prof. Dr. rer. nat. Sebastian Otte - Tue, 14. April 2026, 9:30

Towards Flexible and Efficient Learning

As artificial intelligence systems become deeply embedded in everyday life, their energy consumption has emerged as a critical challenge. Training and deploying large-scale models increasingly require enormous computational resources, in some cases approaching the energy demand of entire cities. At the same time, many real-world applications demand learning systems that operate in dynamic, resource-constrained environments. These pressures raise a fundamental question: how can we design AI models that remain powerful while adapting efficiently, in ways more closely aligned with natural intelligence? This talk presents a research trajectory aimed at developing learning systems that combine flexibility across time scales and structures with efficiency in computation, energy, and representation. Drawing inspiration from neuroscience and dynamical systems, the talk centers on recurrent architectures as a core computational principle, reflecting the deeply recurrent organization of the brain across spatial and temporal scales. Selected contributions illustrate this perspective, including minimal and scan-compatible recurrent networks, spiking models with oscillatory dynamics, and recent refinements of foundational components such as activation functions. Together, these works form building blocks for a broader vision of flexible, scalable, and more sustainable learning systems grounded in the temporal structure of the world.

Sebastian Otte is a professor at the Institute for Robotics and Cognitive Systems at the University of Lübeck in Germany and the head of the Research Group for Adaptive Artificial Intelligence. Sebastian and his team focus on developing flexible and adaptive AI systems that can continuously learn and dynamically adapt to changing environments. One of their main goals is the creation of efficient and sustainable AI that can operate effectively despite resource constraints such as limited energy supply, memory, and computational power. The team works in an interdisciplinary research context, combining artificial intelligence, cognitive science, and robotics to develop innovative and sustainable approaches. From 2013 to 2017, Sebastian was a PhD student in AI and robotics at the Eberhard Karls University of Tübingen, Germany. From 2017 to 2023, he conducted postdoctoral research in the Neuro-Cognitive Modeling Group at the University of Tübingen. In 2020, he served as an acting professor for the Chair of Distributed Intelligence at the University of Tübingen. From 2022 to 2023, supported by a Feodor Lynen Research Fellowship from the Alexander von Humboldt Foundation, he worked as a visiting scientist at the Centrum Wiskunde & Informatica (CWI) in Amsterdam, Netherlands. In September 2023, Sebastian assumed his current position as a professor at the University of Lübeck.

Assoc. Prof. Dr. Berk Calli - Tue, 14. April 2026, 11:00

Leveraging Vision, Compliance, and Learning for Robust Robotic Manipulation in Unstructured Environments

Robotic manipulation is essential for enabling robots to operate in unstructured environments. This talk begins with environmental robotics applications such as shipbreaking and recycling sorting, where real-world complexity demands interdisciplinary approaches that integrate robotics, domain knowledge, and system-level design. These applications expose key challenges in perception and physical interaction. Motivated by these challenges, the talk then focuses on fundamental problems in robotic manipulation, particularly grasping and vision-based control, highlighting how tighter integration of mechanism design, perception and control can improve robustness and adaptability.

Berk Calli is an Associate Professor in the Robotics Engineering Department at Worcester Polytechnic Institute (WPI), where he leads the Manipulation and Environmental Robotics (MER) Laboratory. His research focuses on addressing environmental challenges through robotics, alongside fundamental problems in robotic manipulation. His work spans robotic grasping, in-hand manipulation, soft robot control, robot learning, and active vision strategies. He received the U.S. National Science Foundation CAREER Award in 2024. He is a lead organizer of the Climate Robotics Summit, the COMPARE community ecosystem, and the ICRA Robotic Grasping and Manipulation Competition. Prior to joining WPI, he was a postdoctoral researcher at the Yale University GRAB Lab. He received his PhD from Delft University of Technology in the Netherlands and his MS and BS degrees in Mechatronics Engineering from Sabancı University, Turkey.

Dr. techn. Dipl. Ing. Univ. Sebastian Böck, NXAI - Wed, 15. April 2026, 14:00

Beyond Transformers: Linear Scaling, In-Context Learning, and the xLSTM Paradigm

While Transformers dominate modern AI, their quadratic computational complexity and lack of state- tracking present severe bottlenecks. This keynote explores the architectural shift towards the extended Long Short-Term Memory (xLSTM) paradigm, which integrates novel memory structures to deliver linear-time compute scaling, a constant memory footprint, and highly competitive in-context learning. Highlighting recent milestones, the presentation will cover advancements in large language modeling, efficient model distillation, and time series forecasting. The session will conclude by outlining the broad applicability of recurrent backbones across diverse domains—including vision, biology, and robotics—and exploring advanced memory concepts for future Agentic AI.

Sebastian Böck is a Lead Researcher at NXAI, specializing in Deep Learning, Sequence Modeling, and Time Series Forecasting. He received his PhD in Computer Science from JKU in Linz and his diploma degree in Electrical Engineering and Information Technology from the Technical University in Munich. Long before his current focus on the efficiency and scalability of foundation models, he pioneered the application of recurrent neural networks (RNNs) in Music Information Retrieval—spanning onset detection, beat tracking, and polyphonic transcription. This early work cemented his deep expertise in state-tracking and recurrent architectures. Today, his research at NXAI centers on overcoming the fundamental scaling and inference bottlenecks of modern AI through the extended Long Short-Term Memory (xLSTM) paradigm and its application across diverse modalities.