BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Date iCal//NONSGML kigkonsult.se iCalcreator 2.20.2//
METHOD:PUBLISH
X-WR-CALNAME;VALUE=TEXT:Eventi DIAG
BEGIN:VTIMEZONE
TZID:Europe/Paris
BEGIN:STANDARD
DTSTART:20251026T030000
TZOFFSETFROM:+0200
TZOFFSETTO:+0100
TZNAME:CET
END:STANDARD
BEGIN:DAYLIGHT
DTSTART:20250330T020000
TZOFFSETFROM:+0100
TZOFFSETTO:+0200
TZNAME:CEST
END:DAYLIGHT
END:VTIMEZONE
BEGIN:VEVENT
UID:calendar.29384.field_data.0@www.open.diag.uniroma1.it
DTSTAMP:20260404T211204Z
CREATED:20250630T131744Z
DESCRIPTION:AbstractIn this talk we present a approach to evaluating large 
 language models (LLMs) based on the latest EU GPAI Code of Practice. It ex
 plores the mapping between existing technical benchmarks and the risk cate
 gories defined by the Code\, linking model propensities\, capabilities\, a
 nd risk exposure. The intervention outlines how to operationalize safety m
 easures required by the Code\, including robustness tests\, misuse evaluat
 ions\, and systemic risk assessments. Emphasis is placed on translating ab
 stract commitments into measurable technical processes to support complian
 ce and align LLM development with EU regulatory expectations.
DTSTART;TZID=Europe/Paris:20250717T120000
DTEND;TZID=Europe/Paris:20250717T120000
LAST-MODIFIED:20250630T140124Z
LOCATION:Aula Magna\, Diag\, Via Ariosto 25
SUMMARY:Assessing LLMs Risks under AI-Act Code of Practice - Piercosma Bisc
 onti Lucidi
URL;TYPE=URI:http://www.open.diag.uniroma1.it/node/29384
END:VEVENT
END:VCALENDAR