BEGIN:VCALENDAR
VERSION:2.0
PRODID:IEEE vTools.Events//EN
CALSCALE:GREGORIAN
BEGIN:VTIMEZONE
TZID:America/Los_Angeles
BEGIN:DAYLIGHT
DTSTART:20260308T030000
TZOFFSETFROM:-0800
TZOFFSETTO:-0700
RRULE:FREQ=YEARLY;BYDAY=2SU;BYMONTH=3
TZNAME:PDT
END:DAYLIGHT
BEGIN:STANDARD
DTSTART:20261101T010000
TZOFFSETFROM:-0700
TZOFFSETTO:-0800
RRULE:FREQ=YEARLY;BYDAY=1SU;BYMONTH=11
TZNAME:PST
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20260916T002901Z
UID:57FAA825-86BB-4DDB-B25C-583485353F04
DTSTART;TZID=America/Los_Angeles:20261005T173000
DTEND;TZID=America/Los_Angeles:20261005T200000
DESCRIPTION:Large language models (LLMs) and Generative AI (GenAI) are at t
 he forefront of frontier AI research and technology. With their rapidly in
 creasing popularity and availability\, challenges and concerns about their
  misuse and safety risks are becoming more prominent than ever. In this ta
 lk\, we introduce a unified computational framework for evaluating and imp
 roving a wide range of safety challenges in generative AI. Specifically\, 
 we will show new tools and insights to explore and mitigate the safety and
  robustness risks associated with state-of-the-art LLMs and GenAI models\,
  including (i) safety risks in fine-tuning LLMs\, (ii) LLM red-teaming and
  jailbreak mitigation\, (iii) prompt engineering for safety debugging\, an
 d (iv) robust detection of AI-generated content.\n\nSpeaker(s): Pin-Yu Che
 n\, \n\nAgenda: \n- 5:30 PM – 6:00 PM: Networking (Free pizza & drinks)\
 n- 6:00 PM – 7:30 PM: Presentation\, Q&A\n\nRoom: Bergin Hall Room 116\,
  Bldg: SCDI\, 500 El Camino Real\, Santa Clara\, California\, United State
 s\, 95053
LOCATION:Room: Bergin Hall Room 116\, Bldg: SCDI\, 500 El Camino Real\, San
 ta Clara\, California\, United States\, 95053
ORGANIZER:vickyhlu@ieee.org
SEQUENCE:24
SUMMARY:Computational Safety for Generative AI
URL;VALUE=URI:https://events.vtools.ieee.org/m/577919
X-ALT-DESC:Description: <br /><p class="MsoNormal"><span lang="EN">Large la
 nguage models (LLMs) and Generative AI (GenAI) are at the forefront of fro
 ntier AI research and technology. With their rapidly increasing popularity
  and availability\, challenges and concerns about their misuse and safety 
 risks are becoming more prominent than ever. In this talk\, we introduce a
  unified computational framework for evaluating and improving a wide range
  of safety challenges in generative AI. Specifically\, we will show new to
 ols and insights to explore and mitigate the safety and robustness risks a
 ssociated with state-of-the-art LLMs and GenAI models\, including (i) safe
 ty risks in fine-tuning LLMs\, (ii) LLM red-teaming and jailbreak mitigati
 on\, (iii) prompt engineering for safety debugging\, and (iv) robust detec
 tion of AI-generated content.</span></p><br /><br />Agenda: <br /><ul type
 ="disc">\n<li class="MsoNormal"><strong><span lang="EN">5:30 PM &ndash\; 6
 :00 PM:</span></strong><span lang="EN"> Networking (Free pizza &amp\; drin
 ks)</span></li>\n<li class="MsoNormal"><strong><span lang="EN">6:00 PM &nd
 ash\; 7:30 PM:</span></strong><span lang="EN"> Presentation\, Q&amp\;A</sp
 an></li>\n</ul>
END:VEVENT
END:VCALENDAR