BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//pretalx//pretalx.com//bbuzz22//speaker//ZWB38G
BEGIN:VTIMEZONE
TZID:CET
BEGIN:STANDARD
DTSTART:20001029T040000
RRULE:FREQ=YEARLY;BYDAY=-1SU;BYMONTH=10
TZNAME:CET
TZOFFSETFROM:+0200
TZOFFSETTO:+0100
END:STANDARD
BEGIN:DAYLIGHT
DTSTART:20000326T030000
RRULE:FREQ=YEARLY;BYDAY=-1SU;BYMONTH=3
TZNAME:CEST
TZOFFSETFROM:+0100
TZOFFSETTO:+0200
END:DAYLIGHT
END:VTIMEZONE
BEGIN:VEVENT
UID:pretalx-bbuzz22-7TYXQN@pretalx.com
DTSTART;TZID=CET:20220613T140000
DTEND;TZID=CET:20220613T144000
DESCRIPTION:Semantic search using AI-powered vector embeddings of text\, wh
 ere relevancy is measured using a vector similarity function\, has been a 
 hot topic for the last few years. As a result\, platforms and solutions fo
 r vector search have been springing up like mushrooms. Even traditional se
 arch engines like Elasticsearch and Apache Solr ride the semantic vector s
 earch wave and now support fast but approximative vector search\, a buildi
 ng block for supporting AI-powered semantic search at scale. \n\nUndoublty
 \, sizeable pre-trained language models like BERT have revolutionized the 
 state-of-the-art on data-rich text search relevancy datasets. However\, th
 e question search practitioners are asking themself is\, do these models d
 eliver on their promise of an improved search experience when applied to t
 heir domain? Furthermore\, is semantic search the silver bullet which outc
 ompetes traditional keyword-based search across many search use cases? Thi
 s talk delves into these questions and demonstrates how these semantic mod
 els can dramatically fail to deliver their promise when used on unseen dat
 a in new domains.\n\nThe Search track is presented by OpenSource Connectio
 ns
DTSTAMP:20260515T111055Z
LOCATION:Maschinenhaus
SUMMARY:AI-powered Semantic Search\; A story of broken promises? - Jo Krist
 ian Bergum
URL:https://pretalx.com/bbuzz22/talk/7TYXQN/
END:VEVENT
END:VCALENDAR