BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//pretalx//pretalx.com//bbuzz22//talk//SYJWRU
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-SYJWRU@pretalx.com
DTSTART;TZID=CET:20220613T145000
DTEND;TZID=CET:20220613T153000
DESCRIPTION:### Context\n\nIn the past year the interest in Neural Search a
 nd vector search engines increased heavily. They promise to solve multi mo
 dal\, cross modal and semantic search problems with ease. However\, when q
 uickly trying Neural search with off-the-shelf pre-trained models the resu
 lts are quite disillusioning. They lacking knowledge about the data at han
 d. In order to explicitly solve model finetuning for search problems we im
 plemented an open-source finetuner. It is directly usable with several vec
 tor databases due to the underlying data structure.\n\n### Presentation\n\
 nIn our talk we present our methodology and performance on an example data
 set. Afterwards\, we show how well the approach transfers to other dataset
 s\, such as deepfashion\, geolocation geoguessr and more. It will give han
 ds-on guidance on how you can finetune a model in order to make your data 
 better searchable.\n\nThe Search track is presented by OpenSource Connecti
 ons
DTSTAMP:20260421T125013Z
LOCATION:Frannz Salon
SUMMARY:Neural Search - Let's talk about quality - Maximilian Werk\, Floria
 n Hoenicke
URL:https://pretalx.com/bbuzz22/talk/SYJWRU/
END:VEVENT
END:VCALENDAR