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:20221030T030000
TZOFFSETFROM:+0200
TZOFFSETTO:+0100
TZNAME:CET
END:STANDARD
BEGIN:DAYLIGHT
DTSTART:20230326T020000
TZOFFSETFROM:+0100
TZOFFSETTO:+0200
TZNAME:CEST
END:DAYLIGHT
END:VTIMEZONE
BEGIN:VEVENT
UID:calendar.25678.field_data.0@www.glad.uniroma1.it
DTSTAMP:20260405T091524Z
CREATED:20230317T125208Z
DESCRIPTION:Carissime/i\,vi invito a partecipare al seguente seminario\, in
  cui Edoardo Barba illustrerà la propria attività di ricerca\, in ottemper
 anza ai requisiti previsti dalla procedura valutativa per n.3 posti di Ric
 ercatore a Tempo Determinato tipologia A - SC 09/H1 SSD ING-INF/05 - Dipar
 timento di Ingegneria Informatica Automatica e Gestionale 'A. Ruberti'\, 2
 023RTDAPNRR001\, pubblicato sulla Gazzetta Ufficiale N. 5 del 20.01.2023. 
 Il seminario sarà fruibile anche in modalità telematica via Zoom all'indir
 izzo: https://uniroma1.zoom.us/j/97615093680?pwd=R3U0bDM3aW1qVHM3TGsxYkd0O
 Tc3Zz09Grazie\,Roberto Title: Neuro-symbolic approaches: towards safe\, in
 terpretable and explainable NLPAbstractLarge language models (LLMs) have b
 ecome central to virtually every natural language processing (NLP) task. T
 hey are utilized in a wide range of everyday applications\, including chat
 bots and translation systems\, by millions of users worldwide. Big-tech co
 mpanies and major research groups are investing significant resources in c
 reating more powerful LLMs by using large amounts of data and increasing t
 he number of parameters. However\, this blind pursuit of more powerful mod
 els has come at a cost. This seminar will discuss the environmental impact
 \, biases\, and limitations of large LLMs\, as well as the difficulty of i
 nteracting with them and explaining their behaviour. Most importantly\, we
  will explore a possible solution to these challenges by introducing a nov
 el paradigm for Neuro-symbolic approaches. We will present the main charac
 teristics of this paradigm and explain why it tackles most of the challeng
 es previously introduced. Ultimately\, this seminar will argue that the fu
 ture of NLP lies in the development of more interpretable and transparent 
 models\, capable of taking into account human input and feedback\, and tha
 t can be used to solve real-world problems in a responsible and ethical ma
 nner.Bio: Edoardo Barba is a postdoc researcher at the Department of Compu
 ter\, Control and Management Engineering (DIAG)\, Sapienza University of R
 ome\, working in the area of Natural Language Processing in the Sapienza N
 LP Group led by prof. Roberto Navigli. His interests are in lexical semant
 ics\, sentence-level semantics and retrieval-augmented neuro-symbolic gene
 rative approaches. His research has been published in conferences such as 
 the ACL\, EMNLP\, IJCAI and AAAI.
DTSTART;TZID=Europe/Paris:20230320T120000
DTEND;TZID=Europe/Paris:20230320T120000
LAST-MODIFIED:20230406T061000Z
LOCATION:Aula Magna DIAG
SUMMARY:Seminario pubblico di Edoardo Barba (procedura valutativa per n.3 p
 osti di Ricercatore a tempo determinato tipologia A - SC 09/H1\, SSD ING-I
 NF/05) - Edoardo Barba
URL;TYPE=URI:http://www.glad.uniroma1.it/node/25678
END:VEVENT
END:VCALENDAR