BEGIN:VCALENDAR VERSION:2.0 PRODID:IEEE vTools.Events//EN CALSCALE:GREGORIAN BEGIN:VTIMEZONE TZID:America/New_York BEGIN:DAYLIGHT DTSTART:20240310T030000 TZOFFSETFROM:-0500 TZOFFSETTO:-0400 RRULE:FREQ=YEARLY;BYDAY=2SU;BYMONTH=3 TZNAME:EDT END:DAYLIGHT BEGIN:STANDARD DTSTART:20231105T010000 TZOFFSETFROM:-0400 TZOFFSETTO:-0500 RRULE:FREQ=YEARLY;BYDAY=1SU;BYMONTH=11 TZNAME:EST END:STANDARD END:VTIMEZONE BEGIN:VEVENT DTSTAMP:20240314T200107Z UID:7A1C59EE-0DEA-4B39-BF39-2772C1D8DFD3 DTSTART;TZID=America/New_York:20240307T180000 DTEND;TZID=America/New_York:20240307T190000 DESCRIPTION:Special Presentation on “LMMs as Universal Foundation Models for AI-Native Wireless Systems”\n\nby Dr. Christo K. Thomas (Virginia Te ch\, USA)\n\nHosted by Future Networks Artificial Intelligence & Machine L earning (AIML) Working Group\n\nDate/Time: Thursday\, March 7th\, 2024 @ 6 PM EST (3 PM PST)\n\nTopic:\n\nLarge Multi-Modal Models (LMMs) as Univers al Foundation Models for AI-Native Wireless Systems\n\nAbstract:\n\nFounda tion models such as large language models (LLMs) have recently been touted as game-changers for 6G systems. However\, previous efforts on LLMs for w ireless networks are limited to directly applying existing language models designed for natural language processing (NLP) applications. Contrary to this\, in this talk\, we present a comprehensive vision of how to design u niversal foundation models that are tailored towards the unique needs of n ext-generation wireless systems\, thereby paving the way towards the deplo yment of artificial intelligence (AI)-native networks. These LMMs are driv en by three distinct characteristics: 1) integration of multi-modal sensin g data\, 2) grounding sensory input via causal reasoning and retrieval-aug mented generation (RAG)\, and 3) instructibility to environmental feedback through logical and mathematical reasoning enabled by neuro-symbolic AI. These attributes are crucial for developing "universal foundation models" capable of addressing interconnected cross-layer networking challenges in AI-native wireless systems while ensuring alignment of objectives across d iverse domains. We also discuss preliminary results from experimental eval uation that demonstrate the efficacy of grounding using RAG in LMMs\, and showcase the alignment of LMMs with wireless system designs. Furthermore\, compared to vanilla LLMs\, the enhanced rationale exhibited in the respon ses to mathematical questions by LMMs demonstrates the logical and mathema tical reasoning capabilities inherent in LMMs. Building on those results\, we present a sequel of open questions and challenges for LMMs\, including intent-based networks\, resilient wireless systems\, semantic communicati ons\, and many more.\n\nCo-sponsored by: IEEE Future Networks\n\nSpeaker(s ): Dr. Christo K. Thomas \n\nVirtual: https://events.vtools.ieee.org/m/407 262 LOCATION:Virtual: https://events.vtools.ieee.org/m/407262 ORGANIZER:c.polk@comsoc.org SEQUENCE:12 SUMMARY:LMMs for AI-Native Wireless Systems - AI/ML webinar URL;VALUE=URI:https://events.vtools.ieee.org/m/407262 X-ALT-DESC:Description: <br /><p>Special Presentation on &ldquo\;<strong>LM Ms as Universal Foundation Models for AI-Native Wireless Systems</strong>& rdquo\;</p>\n<p>by<strong> Dr. Christo K. Thomas (Virginia Tech\, USA)</st rong></p>\n<p>Hosted by Future Networks<strong> Artificial Intelligence &a mp\; Machine Learning (AIML) Working Group</strong></p>\n<p><strong>Date/T ime</strong>: <strong>Thursday\, March 7<sup>th</sup>\, 2024 @ 6 PM EST (3 PM PST)</strong></p>\n<p><strong><u>Topic</u></strong><strong>:</strong>< /p>\n<p><strong>Large Multi-Modal Models (LMMs) as Universal Foundation Mo dels for AI-Native Wireless Systems</strong></p>\n<p><strong><u>Abstract</ u></strong><strong>:</strong></p>\n<p>Foundation models such as large lang uage models (LLMs) have recently been touted as game-changers for 6G syste ms. However\, previous efforts on LLMs for wireless networks are limited t o directly applying existing language models designed for natural language processing (NLP) applications. Contrary to this\, in this talk\, we prese nt a comprehensive vision of how to design universal foundation models tha t are tailored towards the unique needs of next-generation wireless system s\, thereby paving the way towards the deployment of artificial intelligen ce (AI)-native networks. These LMMs are driven by three distinct character istics: 1) integration of multi-modal sensing data\, 2) grounding sensory input via causal reasoning and retrieval-augmented generation (RAG)\, and 3) instructibility to environmental feedback through logical and mathemati cal reasoning enabled by neuro-symbolic AI. These attributes are crucial f or developing "universal foundation models" capable of addressing intercon nected cross-layer networking challenges in AI-native wireless systems whi le ensuring alignment of objectives across diverse domains. We also discus s preliminary results from experimental evaluation that demonstrate the ef ficacy of grounding using RAG in LMMs\, and showcase the alignment of LMMs with wireless system designs. Furthermore\, compared to vanilla LLMs\, th e enhanced rationale exhibited in the responses to mathematical questions by LMMs demonstrates the logical and mathematical reasoning capabilities i nherent in LMMs. Building on those results\, we present a sequel of open q uestions and challenges for LMMs\, including intent-based networks\, resil ient wireless systems\, semantic communications\, and many more.</p> END:VEVENT END:VCALENDAR