BEGIN:VCALENDAR
VERSION:2.0
PRODID:www.dresden-science-calendar.de
METHOD:PUBLISH
CALSCALE:GREGORIAN
X-MICROSOFT-CALSCALE:GREGORIAN
X-WR-TIMEZONE:Europe/Berlin
BEGIN:VTIMEZONE
TZID:Europe/Berlin
X-LIC-LOCATION:Europe/Berlin
BEGIN:DAYLIGHT
TZNAME:CEST
TZOFFSETFROM:+0100
TZOFFSETTO:+0200
DTSTART:19810329T030000
RRULE:FREQ=YEARLY;INTERVAL=1;BYMONTH=3;BYDAY=-1SU
END:DAYLIGHT
BEGIN:STANDARD
TZNAME:CET
TZOFFSETFROM:+0200
TZOFFSETTO:+0100
DTSTART:19961027T030000
RRULE:FREQ=YEARLY;INTERVAL=1;BYMONTH=10;BYDAY=-1SU
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
UID:DSC-23046
DTSTART;VALUE=DATE:20260903
SEQUENCE:1784634177
TRANSP:OPAQUE
DTEND;VALUE=DATE:20260905
URL:https://www.dresden-science-calendar.de/calendar/de/detail/23046
LOCATION:Andere\,   
SUMMARY:Razniewski: Resilient AI: Securing Large Language Models
CLASS:PUBLIC
DESCRIPTION:Speaker: Prof. Simon Razniewski\, Dr Himanshu Beniwal\, Dr. Par
 amita Mirza\, Dr. Siavash Ghiasvand\, Dr. Hermann Diebel-Fischer \nInstitu
 te of Speaker: ScaDS.AI Dresden/Leipzig\nTopics:\n\nTime:\n09:00-16:00\n\n
  Location:\n  Name: Andere (Strehlener Straße 14\, 01069 Dresden)\n  Stre
 et:  \n  City:  \n  Phone: \n  Fax: \nDescription: Large Language Models (
 LLMs) are increasingly deployed in research and real-world applications\, 
 where they interact with sensitive data\, external tools\, and human users
 . Despite their capabilities\, LLMs introduce novel security and privacy c
 hallenges\, including prompt injection\, data leakage\, and unintended dis
 closure of sensitive information. These risks stem from the fundamental de
 sign of LLMs\, which interpret natural language instructions without relia
 ble mechanisms to distinguish trusted from adversarial inputs. As a result
 \, ensuring the resilience\, robustness\, and trustworthiness of LLM-based
  systems has become a critical research challenge.    This workshop aims t
 o provide an accessible overview of how LLMs work\, introduce common attac
 k scenarios\, discuss privacy risks\, and present methods for adding safeg
 uards and protective mechanisms when building LLM-based systems.
X-MICROSOFT-CDO-ALLDAYEVENT:TRUE
DTSTAMP:20260911T141602Z
CREATED:20260721T114257Z
LAST-MODIFIED:20260721T114257Z
END:VEVENT
END:VCALENDAR