BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//pretalx//pretalx.com//hack-lu-2024//speaker//CE8DBB
BEGIN:VTIMEZONE
TZID:Europe/Luxembourg
BEGIN:DAYLIGHT
DTSTART:20231023T000000
TZNAME:CEST
TZOFFSETFROM:+0200
TZOFFSETTO:+0200
END:DAYLIGHT
BEGIN:STANDARD
DTSTART:20231029T030000
RDATE:20241027T030000
TZNAME:CET
TZOFFSETFROM:+0200
TZOFFSETTO:+0100
END:STANDARD
BEGIN:DAYLIGHT
DTSTART:20240331T030000
RDATE:20250330T030000
TZNAME:CEST
TZOFFSETFROM:+0100
TZOFFSETTO:+0200
END:DAYLIGHT
END:VTIMEZONE
BEGIN:VEVENT
SUMMARY:NeuroCTI - a custom LLM for CTI - benchmarking\, successes\, failu
 res and lessons learned (updates) - Aaron Kaplan
DTSTART;TZID=Europe/Luxembourg:20241022T093000
DTEND;TZID=Europe/Luxembourg:20241022T100000
DTSTAMP:20261118T235859Z
UID:pretalx-hack-lu-2024-JJVXKP@pretalx.com
DESCRIPTION:LLMs turn out to be highly practical for summarising and extra
 cting information from unstructured Cyber Threat Intelligence (CTI) report
 s. However\, most models were not trained specifically for understanding C
 TI. We will present a custom LLM\, fine-tuned for CTI purposes. But of cou
 rse\, that only makes sense with a CTI text benchmark dataset. Creating th
 ese two systems is a challenging journey. Set-backs guaranteed. We will sh
 are our findings.
LOCATION:Europe - Main Room
URL:https://pretalx.com/hack-lu-2024/talk/JJVXKP/
END:VEVENT
END:VCALENDAR
