Zobrazeno 1 - 10
of 10
pro vyhledávání: '"Matej Ulčar"'
Autor:
Matej Ulčar, Marko Robnik-Šikonja
Publikováno v:
Frontiers in Artificial Intelligence, Vol 6 (2023)
IntroductionLarge pretrained language models have recently conquered the area of natural language processing. As an alternative to predominant masked language modeling introduced in BERT, the T5 model has introduced a more general training objective,
Externí odkaz:
https://doaj.org/article/52757f1102d44804861f566763d0faf5
Publikováno v:
Slovenščina 2.0: Empirične, aplikativne in interdisciplinarne raziskave, Vol 9, Iss 1 (2021)
In recent years, the use of deep neural networks and dense vector embeddings for text representation have led to excellent results in the field of computational understanding of natural language. It has also been shown that word embeddings often capt
Externí odkaz:
https://doaj.org/article/9d4e22fbe85e43159fb32dc66266b2c4
Autor:
Matej Ulčar, Marko Robnik-Šikonja
Publikováno v:
Lecture Notes in Computer Science ISBN: 9783031164996
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::641abc5c15e10515e661c1e46d93b440
https://doi.org/10.1007/978-3-031-16500-9_14
https://doi.org/10.1007/978-3-031-16500-9_14
Publikováno v:
Slovenščina 2.0: Empirične, aplikativne in interdisciplinarne raziskave, Vol 9, Iss 1 (2021)
Slovenščina 2.0: empirical, applied and interdisciplinary research
Slovenščina 2.0: empirical, applied and interdisciplinary research
In recent years, the use of deep neural networks and dense vector embeddings for text representation have led to excellent results in the field of computational understanding of natural language. It has also been shown that word embeddings often capt
Autor:
Senja Pollak, Matthew Purver, Nikola Ljubešić, Mohammad Taher Pilehvar, Ivan Vulić, Matej Ulčar, Carlos Santos Armendariz
Publikováno v:
SemEval@COLING
This paper presents the Graded Word Similarity in Context (GWSC) task which asked participants to predict the effects of context on human perception of similarity in English, Croatian, Slovene and Finnish. We received 15 submissions and 11 system des
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::cbc11035b6b4cfeddde7dfa9ae10c019
Autor:
Marko Robnik-Šikonja, Matej Ulčar
Publikováno v:
Text, Speech, and Dialogue-23rd International Conference, TSD 2020, Brno, Czech Republic, September 8–11, 2020, Proceedings
Text, Speech, and Dialogue ISBN: 9783030583224
TDS
Text, Speech, and Dialogue ISBN: 9783030583224
TDS
Large pretrained masked language models have become state-of-the-art solutions for many NLP problems. The research has been mostly focused on English language, though. While massively multilingual models exist, studies have shown that monolingual mod
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::2a308fec22ba03acf95560ee91643aed
http://arxiv.org/abs/2006.07890
http://arxiv.org/abs/2006.07890
Autor:
Carlos Santos Armendariz, Matthew Purver, Matej Ulčar, Senja Pollak, Nikola Ljubešič, Marko Robnik-Šikonja, Mark Granroth-Wilding, Kristiina Vaik
Publikováno v:
Proceedings of the 12th Conference on Language Resources and Evaluation (LREC 2020)
State of the art natural language processing tools are built on context-dependent word embeddings, but no direct method for evaluating these representations currently exists. Standard tasks and datasets for intrinsic evaluation of embeddings are base
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::2cf2aa091b6f4d44c591dbded6c64133
https://doi.org/10.5281/zenodo.3894565
https://doi.org/10.5281/zenodo.3894565
Publikováno v:
Uporabna informatika (Ljubljana)
V zadnjem času se na področju samodejnega razpoznavanja govora uveljavljajo globoke nevronske mreže, ki nadomeščajo akustično modeliranje z uporabo HMM in GMM modelov ter n-grame za jezikovni model. Za razpoznavanje govorjene slovenščine smo
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::559c0811fd8c9fbddd22ccfa67158817
http://www.dlib.si/details/URN:NBN:SI:doc-WV49C0GL
http://www.dlib.si/details/URN:NBN:SI:doc-WV49C0GL
Autor:
Matej Ulčar, Marko Robnik-Šikonja
Publikováno v:
Neural Computing and Applications.
Building machine learning prediction models for a specific NLP task requires sufficient training data, which can be difficult to obtain for less-resourced languages. Cross-lingual embeddings map word embeddings from a less-resourced language to a res