ICSMACS'26
PLENARY SESSION
Prof Cherif Chibane , CTO AuresTech Inc, USA
Prof. Chibane is a lecturer, Research Scientist, Technologist and Entrepreneurs with more than 30 years of experience in advanced technology development for NASA, DARPA and major Aerospace companies in the US. He is currently the founder and chief scientist of AuresTech, which performs research for the aerospace industry in the US. His current research involves developing and Muli-agent algorithms that enable a collective of flying robots to operate in a SWARM like configuration. Previously, he was a research scientist at MIT – Lincoln Laboratory and Draper Laboratory where he led research in advanced space communications, as well as in aerospace in guidance and navigation. He was one of the early adaptors of configurable computing that is currently being applied configurable computing to Artificial Intelligence (AI) and Machine Learning (ML). He is a former adjunct professor at Fairleigh Dickinson University where he lectured on advanced engineering topics and mentored and advised students at the undergraduate and graduate levels. Prof. Chibane is the holder of many Patents and industry awards and has recently been awarded a key patent in RF energy to DC conversion that is poised to revolutionize all future battery-operated hand-held devices, cars and IoT.
TITLE : Drone Swarm Technology : Evolution, Trends, and the Future of Collective Intelligence
ABSTRACT
Drone swarm technology represents a paradigm shift in autonomous systems, moving away from single, isolated Unmanned Aerial Vehicles (UAVs) toward highly coordinated, decentralized networks. Inspired by biological phenomena like bird flocking and insect swarming, these systems rely on collective intelligence, where individual drones communicate peer-to-peer to execute complex tasks without a central controller. By leveraging advanced edge computing, mesh networking, and robust swarm algorithms, today’s drone swarms possess unprecedented scalability and resilience. If a single unit fails or is intercepted, the remaining swarm dynamically adapts to redistribute the workload, making them invaluable for high-stakes environments ranging from military defense to search-and-rescue operations.
In this talk, Prof. Chibane will demystify the science behind these autonomous fleets by walking us through his team’s latest research and testing results. Attendees will get an inside look at actual performance data and field-test results, revealing the reaming challenges that need to be overcome to make this technology a reality. Finally, we will look ahead at where this technology is sprinting next—from cutting-edge « cross-domain swarms » that blend air, land, and sea robots, to the engineering puzzles scientists are still trying to solve. Whether you are an industry researcher, professional or just curious about the future of robotics, you will walk away with a clear, foundational understanding of the next frontier in mass autonomy.
Prof. Kamel Smaili , LORIA LAB, University of Lorraine
Prof. Kamel Smaïli is Professor of Computer Science at the University of Lorraine, where he specializes in natural language processing (NLP) and conducts research at the LORIA, a joint research laboratory affiliated with the CNRS. He received his engineering degree in computer science from USTHB (Algiers) in 1986 before continuing his academic training in France, where he earned a PhD from Henri Poincaré University in 1991 and later obtained his Habilitation to Supervise Research (HDR) from Nancy 2 University in 2001. From 1999 to 2001, he was seconded to the CNRS, further strengthening his research profile. In 2014, he founded the SMarT research team at LORIA, with a focus on statistical and neural approaches to NLP, as well as deep learning methods for textual and numerical data. Strongly committed to doctoral training and academic mentorship, he has supervised 21 PhD dissertations, including 18 successfully completed and 3 currently in progress. An internationally recognized expert in his field, he has contributed to the evaluation of numerous national and international research programs, including projects funded by ANR, ANRT, and initiatives such as NeuroInsight 2022. He also played a key role in fostering international scientific collaboration through the co-direction of the DataNet International Associated Laboratory (LIA) (2015–2023), a partnership involving Moroccan institutions, including UIR. In addition, he serves on the scientific council of the Centre for Scientific and Technical Research for the Development of the Arabic Language (ALGERIA). With more than 150 scientific publications to his name, he currently coordinates the ANR TRADEF project and previously led the European Chist-Era AMIS project, completed in 2019 and he is involved in a new Chist-Era project (2026-2029). His longstanding engagement with the scientific community is reflected in his leadership in organizing major conferences, including ICNLSSP 2017 and ICALP 2019, as well as his active involvement in the program committees of internationally renowned conferences such as ICASSP, Interspeech, and LREC. A frequent invited speaker in Japan, France, the Maghreb, and across Europe, he has also participated in more than 80 PhD and HDR examination committees worldwide.
TITLE : From Judge to Mentor: How LLMs Can Guide Small Models to Detect Fake News
ABSTRACT
This presentation explores the role of large language models (LLMs) in fake news detection. While LLMs like GPT-3.5 and GPT-4 show some ability to detect fake news through prompting (zero-shot, few-shot, chain-of-thought), they still underperform compared to task-specific fine-tuned small language models (SLMs) such as BERT. Hu et al. propose that instead of using LLMs as direct detectors, they can serve as advisors, providing rich rationales (textual and commonsense) that help SLMs make better decisions. They introduce the Adaptive Rationale Guidance (ARG) network, which enables SLMs to interact with and adaptively select useful LLM generated rationales. To reduce inference cost, they also propose ARG-D, a distilled version that no longer requires LLM queries at test time. Experiments on Chinese and English datasets show that ARG outperforms baseline SLMs and even some state-of-the-art models, while ARG-D maintains much of that improvement without LLM dependence.
Prof Ruggero Donida Labati University degli Studi di Milano, Italy
Ruggero Donida Labati received the Ph.D. degree in computer science from the Università degli Studi di Milano, Italy, in 2013. Since 2022, he has been an Associate Professor with the Università degli Studi di Milano. He has been a Visiting Researcher with Michigan State University, MI, USA. His original results have been published in more than 90 papers in international journals, proceedings of international conferences, books, and book chapters. His research interests include biometric systems, artificial intelligence and machine learning, signal and image processing, pattern analysis and recognition, and theory and industrial applications of neural networks. He is Chair of the EURASIP BForSec Technical Area Committee. Dr. Donida Labati has been an Associate Editor of Journal of Ambient Intelligence and Humanized Computing (Springer) and is currently an Associate Editor of the EURASIP Journal on Information Security (Springer).
TITLE : Artificial Intelligence for Biometric Applications
ABSTRACT
The number of biometric applications and devices is continuously growing on global scale and biometrics is pervasively entering the everyday life of users. This relevant expansion is producing new challenges and requirements to be fulfilled by the designers. Features such as adaptability, enhanced interactions with the user, robustness to non-ideal conditions, real-time capability, privacy-compliance, and high accuracy are strongly required in innovative applications and solutions, such as: cyber security, smart devices, and ambient intelligent infrastructures. The talk will focus on innovative biometric recognition approaches and systems, with specific attention on recent approaches based on artificial intelligence and deep learning techniques. Artificial intelligence methods and deep learning approaches are capable to learn discriminative features directly from complex multidimensional signals and increase the accuracy, adaptability, and robustness to non-ideal conditions of biometric systems with respect to traditional approaches. The talk will present biometric systems from a technological point of view and will provide an excursus of recent artificial intelligence approaches, including deep learning methods with current strong points and limitations.
Prof. Nadia Zenati Research Director, Center for Development of Advanced Technologies (CDTA), Algiers
Dr. Nadia Zenati is a Research Director at the Center for Development of Advanced Technologies (CDTA), Algiers. She obtained her Ph.D. in Computer Science from the University of Franche-Comté, France, in 2008. She leads the Human Interaction System, Virtual Reality and Augmented Reality (IRVA) research team within the Robotics and Industrial Automation Division. Her research interests include Virtual Reality (VR), Augmented Reality (AR), Mixed Reality (MR), Human-Machine Interaction (HMI), Human-Robot Interaction (HRI), Collaborative Virtual Environments (CVE), 3D interaction techniques, multimodal systems, computer vision for human-machine interfaces, and software architectures for interactive systems. Dr. Nadia Zenati has authored and co-authored more than one hundred scientific publications in international journals and conference proceedings and has participated in several national and international research projects. Her research focuses on how virtual and augmented reality can provide innovative solutions in healthcare, particularly in functional rehabilitation and medical training. Her recent work explores the integration of AI and extended reality (XR) technologies for medical diagnosis, healthcare training, rehabilitation, educational applications and environmental monitoring.
TITLE : Immersive technologies in Medical Training and practice: Current Advances and Future trends
ABSTRACT
Immersive technologies such as Virtual Reality (VR), Augmented Reality (AR) and Mixed Reality (MR) are profoundly transforming methods of medical training and clinical practice. Thanks to their ability to simulate and reproduce complex clinical situations in safe, these innovative technologies offer new opportunities to reinforce and improve learning, refine decision-making and enhance the technical skills of healthcare professionals. This conference will provide a state-of-the-art overview of immersive applications in the medical field, highlighting recent advances in some of the most promising areas, including surgery, diagnostic support, functional rehabilitation and medical training. It will also address the technical, ethical, and pedagogical challenges that accompany the integration of these tools into healthcare systems. Finally, future trends will be explored, including Artificial Intelligence, personalized learning experiences, and the interoperability of immersive platforms, which will help shape and redefine the medicine of tomorrow.