BEGIN:VCALENDAR VERSION:2.0 PRODID:IEEE vTools.Events//EN CALSCALE:GREGORIAN BEGIN:VTIMEZONE TZID:Asia/Singapore BEGIN:DAYLIGHT DTSTART:20380119T111407 TZOFFSETFROM:+0800 TZOFFSETTO:+0800 RRULE:FREQ=YEARLY;BYDAY=3TU;BYMONTH=1 TZNAME:+08 END:DAYLIGHT BEGIN:STANDARD DTSTART:19820101T000000 TZOFFSETFROM:+0730 TZOFFSETTO:+0800 RRULE:FREQ=YEARLY;BYDAY=1FR;BYMONTH=1 TZNAME:+08 END:STANDARD END:VTIMEZONE BEGIN:VEVENT DTSTAMP:20260930T064557Z UID:609DAE0D-C417-41E1-9C33-B0D9EE8ED270 DTSTART;TZID=Asia/Singapore:20261013T093000 DTEND;TZID=Asia/Singapore:20261013T233000 DESCRIPTION:Title: From Large Language Models to Large Perceptive Models: A I That Understands the Physical World\n\nDate: 13th Oct 2026 (Tuesday)\n\n Time: 0930 – 1130 am\n\nVenue: EEE Executive Seminar Rm (S2.2-B2-53)\n\n Abstract:\nLarge Language Models have demonstrated that sufficiently large models trained on massive corpora can acquire remarkable capabilities for understanding\, reasoning\, and generation. Yet language represents only a small fraction of the information generated by the physical world. Futur e intelligent systems will increasingly need to understand not only text\, but also radio signals\, images\, video\, radar\, LiDAR\, acoustic signal s\, wireless channels\, and other sensory modalities. This talk introduces the concept of Large Perceptive Models (LPMs): foundation models designed to learn general-purpose representations of the physical world from heter ogeneous sensing data. Rather than developing a separate AI model for ever y sensor\, environment\, and task\, LPMs aim to learn transferable represe ntations that can support perception\, localization\, sensing\, prediction \, communication\, and decision-making across multiple domains. Particular attention will be given to wireless and RF signals\, where the environmen t itself becomes a source of information about objects\, people\, mobility \, materials\, and electromagnetic propagation.\n\nSpeaker: Prof. Merouane Debbah\, Senior Director\, Digital Future Institute\, Khalifa University\ , Abu Dhabi\n\nBio: Merouane Debbah is a researcher\, educator and technol ogy entrepreneur. Over his career\, he has founded several public and indu strial research centers\, start-ups and is now Professor at Khalifa Univer sity of Science and Technology in Abu Dhabi and founding Senior Director o f KU Digital Future Institute. He is a frequent keynote speaker at interna tional events in the field of telecommunication and AI. His research has b een lying at the interface of fundamental mathematics\, algorithms\, stati stics\, information and communication sciences with a special focus on ran dom matrix theory and learning algorithms. In the Communication field\, he has been at the heart of the development of small cells (4G)\, Massive MI MO (5G) and Large Intelligent Surfaces (6G) technologies. In the AI field\ , he is known for his work on Large Language Models\, distributed AI syste ms for networks and semantic communications. He received multiple prestigi ous distinctions\, prizes and best paper awards (more than 50 IEEE best pa per awards) for his contributions to both fields and according to research .com is ranked as the best scientist in France in the field of Electronics and Electrical Engineering. He is an IEEE Fellow\, a WWRF Fellow\, a Eura sip Fellow\, an AAIA Fellow\, an Institut Louis Bachelier Fellow\, an AIIA Fellow and a Membre émérite SEE. More recently\, his work has focused o n the convergence of generative AI and telecommunications. He has contribu ted to the development of domain-specific foundation models for networks\, including TelecomGPT\, a family of large language models designed to unde rstand and reason about telecommunications problems. His research also exp lores agentic AI and multi-agent systems for autonomous networks\, RF foun dation models\, Large Perceptive Models\, telecom world models\, semantic communications\, and the use of generative AI for sensing\, network operat ion\, optimization\, and reasoning over the physical world. He is actually chair of the IEEE Large Generative AI Models in Telecom (GenAINet) Emergi ng Technology Initiative and a member of the Marconi Prize Selection Advis ory Committee.\n\nEEE Executive Seminar Rm (S2.2-B2-53) \, Singapore\, Si ngapore\, Singapore LOCATION:EEE Executive Seminar Rm (S2.2-B2-53) \, Singapore\, Singapore\, Singapore ORGANIZER:chau.yuen@ntu.edu.sg SEQUENCE:7 SUMMARY:From Large Language Models to Large Perceptive Models: AI That Unde rstands the Physical World URL;VALUE=URI:https://events.vtools.ieee.org/m/580965 X-ALT-DESC:Description: <br /><p class="MsoNormal"><strong>Title:</strong> <strong>From Large Language Models to Large Perceptive Models: AI That Und erstands the Physical World</strong></p>\n<p class="MsoNormal">&nbsp\;</p> \n<p class="MsoNormal"><strong>Date: 13th Oct 2026 (Tuesday)</strong></p>\ n<p class="MsoNormal"><strong>Time: 0930 &ndash\; 1130 am</strong></p>\n<p class="MsoNormal"><strong>Venue: EEE Executive Seminar Rm (S2.2-B2-53)<sp an style="mso-spacerun: yes\;">&nbsp\; </span></strong></p>\n<p class="Mso Normal">&nbsp\;</p>\n<p class="MsoNormal"><strong>Abstract:</strong><br>La rge Language Models have demonstrated that sufficiently large models train ed on massive corpora can acquire remarkable capabilities for understandin g\, reasoning\, and generation. Yet language represents only a small fract ion of the information generated by the physical world. Future intelligent systems will increasingly need to understand not only text\, but also rad io signals\, images\, video\, radar\, LiDAR\, acoustic signals\, wireless channels\, and other sensory modalities. This talk introduces the concept of Large Perceptive Models (LPMs): foundation models designed to learn gen eral-purpose representations of the physical world from heterogeneous sens ing data. Rather than developing a separate AI model for every sensor\, en vironment\, and task\, LPMs aim to learn transferable representations that can support perception\, localization\, sensing\, prediction\, communicat ion\, and decision-making across multiple domains. Particular attention wi ll be given to wireless and RF signals\, where the environment itself beco mes a source of information about objects\, people\, mobility\, materials\ , and electromagnetic propagation.</p>\n<p class="MsoNormal">&nbsp\;</p>\n <p class="MsoNormal"><strong>Speaker</strong>: Prof. Merouane Debbah\, Sen ior Director\, Digital Future Institute\, Khalifa University\, Abu Dhabi</ p>\n<p class="MsoNormal">&nbsp\;</p>\n<p class="MsoNormal"><strong>Bio:</s trong> Merouane Debbah is a researcher\, educator and technology entrepren eur. Over his career\, he has founded several public and industrial resear ch centers\, start-ups and is now Professor at&nbsp\; Khalifa University o f Science and Technology in Abu Dhabi and founding Senior Director of KU D igital Future Institute. He is a frequent keynote speaker at international events in the field of telecommunication and AI. His research has been ly ing at the interface of fundamental mathematics\, algorithms\, statistics\ , information and communication sciences with a special focus on random ma trix theory and learning algorithms. In the Communication field\, he has b een at the heart of the development of small cells (4G)\, Massive MIMO (5G ) and Large Intelligent Surfaces (6G) technologies. In the AI field\, he i s known for his work on Large Language Models\, distributed AI systems for networks and semantic communications. He received multiple prestigious di stinctions\, prizes and best paper awards (more than 50 IEEE best paper aw ards) for his contributions to both fields and according to research.com i s ranked as the best scientist in France in the field of Electronics and E lectrical Engineering. He is an IEEE Fellow\, a WWRF Fellow\, a Eurasip Fe llow\, an AAIA Fellow\, an Institut Louis Bachelier Fellow\, an AIIA Fello w&nbsp\; and a Membre &eacute\;m&eacute\;rite SEE. More recently\, his wor k has focused on the convergence of generative AI and telecommunications. He has contributed to the development of domain-specific foundation models for networks\, including TelecomGPT\, a family of large language models d esigned to understand and reason about telecommunications problems. His re search also explores agentic AI and multi-agent systems for autonomous net works\, RF foundation models\, Large Perceptive Models\, telecom world mod els\, semantic communications\, and the use of generative AI for sensing\, network operation\, optimization\, and reasoning over the physical world. He is actually chair of&nbsp\; the IEEE Large Generative AI Models in Tel ecom (GenAINet) Emerging Technology Initiative and&nbsp\; a member of the Marconi Prize Selection Advisory Committee.</p>\n<p class="MsoNormal">&nbs p\;</p> END:VEVENT END:VCALENDAR