BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//CERN//INDICO//EN
BEGIN:VEVENT
SUMMARY:Implementacija orodij za jezikovne naloge z uporabo programskih vm
 esnikov do velikih jezikovnih modelov | Implementation of API-based LLM to
 ols for language tasks with evaluation
DTSTART:20261126T090000Z
DTEND:20261126T130000Z
DTSTAMP:20261114T050300Z
UID:indico-event-4023@indico.ijs.si
CONTACT:events@slaif.si\;mojca.brglez@ijs.si\;spela.vintar@ijs.si
DESCRIPTION:Speakers: Špela Vintar (CMI JSI)\, Mojca Brglez\n\nIzvajalec 
 / Course provider: Jožef Stefan Institute (JSI)\, Network Infrastructure 
 Centre (CMI)Predavatelji / Instructors: Špela Vintar (CMI JSI)\, Mojca Br
 glez (CMI JSI)Learning objectives: Learning about the use of large languag
 e models via application programming interfaces (APIs)\, to gain an unders
 tanding that enables effective planning of automation of language and othe
 r tasks\n \nCourse content: Large language models can perform a whole ran
 ge of language tasks\, such as translation\, revision\, checking of profes
 sional terminology\, summarizing or scraping data from text strings. It is
  often easiest to include a language model in the workflow via an API\, wh
 ich means that a specific language task that we want to perform repeatedly
  or on a larger amount of data can be automated with an appropriate call a
 nd a program interface to the selected non-local model.\nParticipants are 
 first introduced to the technical aspects of accessing models via APIs\, w
 ith practical examples of program scripts\, and also by testing the design
  of calls for different input data formats and types of tasks. This is fol
 lowed by a more in-depth practical part\, where we present three examples 
 of language tasks\, compare different calls\, and the results of the model
 s accessed via the API interface. Participants are encouraged to prepare t
 heir own data before the training\, on which they can test the presented m
 ethods.\nIn the last part of the training\, we talk about evaluating the r
 esults\, which usually includes creating an evaluation set of correct solu
 tions (gold standard) and using various evaluation metrics to assess the p
 erformance of the model.\n \nLearning outcomes: Understanding the technic
 al aspects of accessing language models through programming interfaces\, t
 he ability to plan their own usage scenarios\, prepare data\, and evaluate
  results\n\nhttps://indico.ijs.si/event/4023/
IMAGE;VALUE=URI:https://indico.ijs.si/event/4023/logo-3560331715.png
LOCATION:Soba 38-39 "e-predavalnica" (Teslova)
URL:https://indico.ijs.si/event/4023/
END:VEVENT
END:VCALENDAR