BEGIN:VCALENDAR
VERSION:2.0
PRODID:IEEE vTools.Events//EN
CALSCALE:GREGORIAN
BEGIN:VEVENT
DTSTAMP:20260124T054226Z
UID:39E18774-A48C-45AC-9794-D7CDD87BABB1
DTSTART;TZID=Etc/UTC:20250220T120000
DTEND;TZID=Etc/UTC:20250220T130000
DESCRIPTION:Special Presentation by Dr. Mahdi Boloursaz Mashhadi (U. of Sur
 rey\, UK)\n\nHosted by the Future Networks Artificial Intelligence &amp; Machi
 ne Learning (AIML) Working Group\n\nDate/Time: Thursday\, 20 February 2025
  @ 12:00 UTC\n\nTopic:\n\nGoal-Oriented Generative Semantic Communications
  with Multimodal LLMs\n\nAbstract:\n\nThe integration of Generative Artifi
 cial Intelligence (GenAI) models with wireless networks provides ample opp
 ortunities to develop innovative technologies with transformative potentia
 l. One such technologies is Generative Semantic Communications (Gen SemCom
 )\, which leverages the capabilities of state-of-the-art GenAI models to d
 evelop ultra-low bitrate semantic communication systems aiming to transmit
  only the semantic message of interest with high fidelity. GenAI models su
 ch as Diffusion\, Flow-based\, and GAN models\, can learn the general dist
 ribution of natural signals through training and generate new samples at t
 he inference time. This generative process can be guided or conditioned to
  synthesize outputs with a desired semantic content. In Gen SemCom\, the s
 emantics of interest are extracted at the transmitter\, communicated over 
 the channel\, and then used at the receiver to guide a generative model to
  locally synthesize a semantically consistent signal. The emerging generat
 ive foundation AI models and Multi-modal Large Language Models (MLLMs) can
  be leveraged in the SemCom framework to convey the most important semanti
 cs of the source signal to the receiver through textual prompts in a super
  compact form. These models possess a vast general knowledge through inten
 sive pre-training on huge amount of data. This alleviates the need for a s
 hared knowledge base/graph between the semantic transmitter and receiver\,
  obviating the need for corresponding knowledge sharing overheads imposed 
 in current SemCom frameworks. Despite the above benefits\, deployment of s
 uch large models in the SemCom framework is challenging due to their high 
 computational complexity\, energy consumption\, and latency. This talk foc
 uses on novel generative approaches to semantic communications\, the funda
 mental bounds on Gen SemCom\, and its emerging applications in wireless ne
 tworks. It investigates the challenges and opportunities of deploying Gen 
 SemCom at various layers in future wireless networks and provides the corr
 esponding future research directions.\n\nSpeaker:\n\n[]\nDr. Mahdi Bolours
 az Mashhadi (Senior Member\, IEEE) is a Lecturer at the 5G/6G Innovation C
 entre (5G/6GIC) at the Institute for Communication Systems (ICS)\, Univers
 ity of Surrey (UoS)\, and a Surrey AI fellow. His research is focused at t
 he intersection of AI/ML with wireless communication\, learning and commun
 ication co-design\, generative AI for telecommunications\, and collaborati
 ve machine learning. He received B.S.\, M.S.\, and Ph.D. degrees in mobile
  telecommunications from the Sharif University of Technology (SUT)\, Tehra
 n\, Iran. He has more than 40 peer reviewed publications and patents in th
 e areas of wireless communications\, machine learning\, and signal process
 ing. He is a PI/Co-PI for various government and industry funded projects 
 including the UKTIN/DSIT 12M£ national project TUDOR. He received the Bes
 t Paper Award from the IEEE EWDTS conference\, and the Exemplary Reviewer 
 Award from the IEEE ComSoc in 2021 and 2022. He served as a panel judge fo
 r the International Telecommunication Union (ITU) on the “AI/ML in 5G”
  challenge 2021- 2022. He is an editor for the Springer Nature Wireless Pe
 rsonal Communications Journal.\n\nCo-sponsored by: Future Networks Artific
 ial Intelligence &amp; Machine Learning (AIML) Working Group\n\nVirtual: https
 ://events.vtools.ieee.org/m/463737
LOCATION:Virtual: https://events.vtools.ieee.org/m/463737
ORGANIZER:baw@ieee.org
SEQUENCE:27
SUMMARY:Goal-Oriented Generative Semantic Communications with Multimodal LL
 Ms
URL;VALUE=URI:https://events.vtools.ieee.org/m/463737
X-ALT-DESC:Description: &lt;br /&gt;&lt;p class=&quot;MsoNormal&quot; style=&quot;margin-top: .25in
 \;&quot;&gt;&lt;img src=&quot;https://events.vtools.ieee.org/vtools_ui/media/display/977da
 e11-d174-43e3-9aea-b6aaf6e1a8ce&quot; width=&quot;1050&quot; height=&quot;275&quot;&gt;&lt;/p&gt;\n&lt;p class=
 &quot;MsoNormal&quot; style=&quot;margin-top: 12.0pt\;&quot;&gt;Special Presentation by&lt;strong&gt; D
 r. Mahdi Boloursaz Mashhadi (U. of Surrey\, UK)&lt;/strong&gt;&lt;/p&gt;\n&lt;p class=&quot;Ms
 oNormal&quot; style=&quot;margin-top: 12.0pt\;&quot;&gt;Hosted by the Future Networks&lt;strong
 &gt; Artificial Intelligence &amp;amp\; Machine Learning (AIML) Working Group&lt;/st
 rong&gt;&lt;/p&gt;\n&lt;p class=&quot;MsoNormal&quot; style=&quot;margin-top: 12.0pt\;&quot;&gt;&lt;strong&gt;&lt;span
  style=&quot;font-size: 14.0pt\; font-family: Copperplate\; mso-fareast-font-fa
 mily: PMingLiU\; mso-fareast-theme-font: minor-fareast\; mso-bidi-font-fam
 ily: Arial\; mso-bidi-theme-font: minor-bidi\; mso-ansi-language: EN-US\; 
 mso-fareast-language: ZH-TW\; mso-bidi-language: AR-SA\;&quot;&gt;Date/Time&lt;/span&gt;
 &lt;/strong&gt;&lt;span style=&quot;font-size: 12.0pt\; font-family: &#39;Calibri&#39;\,sans-ser
 if\; mso-ascii-theme-font: minor-latin\; mso-fareast-font-family: PMingLiU
 \; mso-fareast-theme-font: minor-fareast\; mso-hansi-theme-font: minor-lat
 in\; mso-bidi-font-family: Arial\; mso-bidi-theme-font: minor-bidi\; mso-a
 nsi-language: EN-US\; mso-fareast-language: ZH-TW\; mso-bidi-language: AR-
 SA\;&quot;&gt;: &lt;strong&gt;Thursday\, 20 February 2025&lt;/strong&gt;&lt;strong&gt;&amp;nbsp\;@ 12:00
  UTC&lt;/strong&gt;&lt;/span&gt;&lt;/p&gt;\n&lt;p class=&quot;MsoNormal&quot; style=&quot;margin-top: .25in\;&quot;
 &gt;&lt;strong&gt;&lt;u&gt;&lt;span style=&quot;font-size: 16.0pt\; font-family: Copperplate\;&quot;&gt;T
 opic&lt;/span&gt;&lt;/u&gt;&lt;/strong&gt;&lt;strong&gt;&lt;span style=&quot;font-size: 16.0pt\; font-fami
 ly: Copperplate\;&quot;&gt;:&lt;/span&gt;&lt;/strong&gt;&lt;/p&gt;\n&lt;p class=&quot;MsoNormal&quot;&gt;&lt;strong&gt;&lt;sp
 an style=&quot;font-size: 16.0pt\;&quot;&gt;Goal-Oriented Generative Semantic Communica
 tions with Multimodal LLMs&amp;nbsp\;&lt;/span&gt;&lt;/strong&gt;&lt;/p&gt;\n&lt;p class=&quot;MsoNormal
 &quot; style=&quot;margin-top: .25in\;&quot;&gt;&lt;strong&gt;&lt;u&gt;&lt;span style=&quot;font-size: 16.0pt\; 
 font-family: Copperplate\;&quot;&gt;Abstract&lt;/span&gt;&lt;/u&gt;&lt;/strong&gt;&lt;strong&gt;&lt;span styl
 e=&quot;font-size: 16.0pt\; font-family: Copperplate\;&quot;&gt;:&lt;/span&gt;&lt;/strong&gt;&lt;/p&gt;\n
 &lt;p class=&quot;MsoNormal&quot;&gt;The integration of Generative Artificial Intelligence
  (GenAI) models with wireless networks provides ample opportunities to dev
 elop innovative technologies with transformative potential. One such techn
 ologies is &lt;strong&gt;&lt;em&gt;Generative Semantic Communications (Gen SemCom)&lt;/em
 &gt;\,&lt;/strong&gt; which leverages the capabilities of state-of-the-art GenAI mo
 dels to develop ultra-low bitrate semantic communication systems aiming to
  transmit only the semantic message of interest with high fidelity. GenAI 
 models such as &lt;strong&gt;&lt;em&gt;Diffusion\, Flow-based\, and GAN models\,&lt;/em&gt;&lt;
 /strong&gt;&amp;nbsp\;can learn the general distribution of natural signals throu
 gh training and generate new samples at the inference time. This generativ
 e process can be guided or conditioned to synthesize outputs with a desire
 d semantic content. In Gen SemCom\, the semantics of interest are extracte
 d at the transmitter\, communicated over the channel\, and then used at th
 e receiver to guide a generative model to &lt;strong&gt;&lt;em&gt;locally synthesize a
  semantically consistent signal.&lt;/em&gt;&lt;/strong&gt;&amp;nbsp\;The emerging &lt;strong&gt;
 &lt;em&gt;generative foundation AI models and Multi-modal Large Language Models 
 (MLLMs)&lt;/em&gt;&lt;/strong&gt;&amp;nbsp\;can be leveraged in the SemCom framework to co
 nvey the most important semantics of the source signal to the receiver &lt;st
 rong&gt;&lt;em&gt;through textual prompts in a super compact form. &lt;/em&gt;&lt;/strong&gt;Th
 ese models possess a vast general knowledge through intensive pre-training
  on huge amount of data. This alleviates the need for a shared knowledge b
 ase/graph between the semantic transmitter and receiver\, obviating the ne
 ed for corresponding knowledge sharing overheads imposed in current SemCom
  frameworks. Despite the above benefits\, deployment of such large models 
 in the SemCom framework is challenging due to their &lt;strong&gt;&lt;em&gt;high compu
 tational complexity\, energy consumption\, and latency.&lt;/em&gt;&lt;/strong&gt; This
  talk focuses on novel generative approaches to semantic communications\, 
 the fundamental bounds on Gen SemCom\, and its emerging applications in wi
 reless networks. It investigates the challenges and opportunities of deplo
 ying Gen SemCom at various layers in future wireless networks and provides
  the corresponding future research directions.&lt;/p&gt;\n&lt;p&gt;&lt;strong&gt;&lt;span style
 =&quot;font-size: 16.0pt\; font-family: Copperplate\;&quot;&gt;&lt;u&gt;Speaker&lt;/u&gt;:&lt;/span&gt;&lt;/
 strong&gt;&lt;/p&gt;\n&lt;table style=&quot;border-collapse: collapse\; width: 100%\;&quot; bord
 er=&quot;1&quot;&gt;&lt;colgroup&gt;&lt;col style=&quot;width: 21.017274%\;&quot;&gt;&lt;col style=&quot;width: 78.88
 6756%\;&quot;&gt;&lt;/colgroup&gt;\n&lt;tbody&gt;\n&lt;tr&gt;\n&lt;td&gt;&lt;img src=&quot;https://events.vtools.i
 eee.org/vtools_ui/media/display/164d92d1-57f6-468f-8e1d-1f54c9a68579&quot; alt=
 &quot;&quot; width=&quot;190&quot; height=&quot;257&quot;&gt;&lt;/td&gt;\n&lt;td&gt;\n&lt;p class=&quot;MsoNormal&quot; style=&quot;margi
 n-top: 6.0pt\;&quot;&gt;Dr. Mahdi Boloursaz Mashhadi (Senior Member\, IEEE) is a L
 ecturer at the 5G/6G Innovation Centre (5G/6GIC) at the Institute for Comm
 unication Systems (ICS)\, University of Surrey (UoS)\, and a Surrey AI fel
 low. His research is focused at the intersection of AI/ML with wireless co
 mmunication\, learning and communication co-design\, generative AI for tel
 ecommunications\, and collaborative machine learning. He received B.S.\, M
 .S.\, and Ph.D. degrees in mobile telecommunications from the Sharif Unive
 rsity of Technology (SUT)\, Tehran\, Iran. He has more than 40 peer review
 ed publications and patents in the areas of wireless communications\, mach
 ine learning\, and signal processing. He is a PI/Co-PI for various governm
 ent and industry funded projects including the UKTIN/DSIT 12M&amp;pound\; nati
 onal project TUDOR. He received the Best Paper Award from the IEEE EWDTS c
 onference\, and the Exemplary Reviewer Award from the IEEE ComSoc in 2021 
 and 2022. He served as a panel judge for the International Telecommunicati
 on Union (ITU) on the &amp;ldquo\;AI/ML in 5G&amp;rdquo\; challenge 2021- 2022. He
  is an editor for the Springer Nature Wireless Personal Communications Jou
 rnal.&lt;/p&gt;\n&lt;p class=&quot;MsoNormal&quot; style=&quot;margin-top: 6.0pt\;&quot;&gt;&amp;nbsp\;&lt;/p&gt;\n&lt;
 /td&gt;\n&lt;/tr&gt;\n&lt;/tbody&gt;\n&lt;/table&gt;
END:VEVENT
END:VCALENDAR