Zobrazeno 1 - 10
of 52
pro vyhledávání: '"Agić, Željko"'
Autor:
Borgholt, Lasse, Havtorn, Jakob Drachmann, Agić, Željko, Søgaard, Anders, Maaløe, Lars, Igel, Christian
The two most common paradigms for end-to-end speech recognition are connectionist temporal classification (CTC) and attention-based encoder-decoder (AED) models. It has been argued that the latter is better suited for learning an implicit language mo
Externí odkaz:
http://arxiv.org/abs/2102.09928
Autor:
Havtorn, Jakob D., Latko, Jan, Edin, Joakim, Borgholt, Lasse, Maaløe, Lars, Belgrano, Lorenzo, Jacobsen, Nicolai F., Sdun, Regitze, Agić, Željko
We address a challenging and practical task of labeling questions in speech in real time during telephone calls to emergency medical services in English, which embeds within a broader decision support system for emergency call-takers. We propose a no
Externí odkaz:
http://arxiv.org/abs/2005.00812
Current methods of cross-lingual parser transfer focus on predicting the best parser for a low-resource target language globally, that is, "at treebank level". In this work, we propose and argue for a novel cross-lingual transfer paradigm: instance-l
Externí odkaz:
http://arxiv.org/abs/2004.07642
In natural language processing, the deep learning revolution has shifted the focus from conventional hand-crafted symbolic representations to dense inputs, which are adequate representations learned automatically from corpora. However, particularly w
Externí odkaz:
http://arxiv.org/abs/1811.08757
Autor:
Plank, Barbara, Agić, Željko
We introduce DsDs: a cross-lingual neural part-of-speech tagger that learns from disparate sources of distant supervision, and realistically scales to hundreds of low-resource languages. The model exploits annotation projection, instance selection, t
Externí odkaz:
http://arxiv.org/abs/1808.09733
Autor:
Agić, Željko, Schluter, Natalie
The recent years have seen a revival of interest in textual entailment, sparked by i) the emergence of powerful deep neural network learners for natural language processing and ii) the timely development of large-scale evaluation datasets such as SNL
Externí odkaz:
http://arxiv.org/abs/1704.05347
We propose UDP, the first training-free parser for Universal Dependencies (UD). Our algorithm is based on PageRank and a small set of head attachment rules. It features two-step decoding to guarantee that function words are attached as leaf nodes. Th
Externí odkaz:
http://arxiv.org/abs/1701.03163
Publikováno v:
In Procedia - Social and Behavioral Sciences 25 October 2013 95:490-497
Autor:
Agić, Željko
Kontinuirani proizvodni procesi farmaceutika sve su važniji zahvaljujući većoj efikasnosti, smanjenju troškova i boljim osiguranjem kvalitete u odnosu na serijsku proizvodnju. Kako bi kontinuirani proizvodni proces bio valjan nužno je osigurati
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=57a035e5b1ae::8d21fcc1d9532e8af2c2e50a8e5ad19e
https://www.bib.irb.hr/1256564
https://www.bib.irb.hr/1256564
Autor:
Agić, Željko, Schluter, Natalie
Publikováno v:
Agic, Z & Schluter, N 2018, Baselines and test data for cross-lingual inference . in Proceedings of the Eleventh International Conference on Language Resources and Evaluation (LREC 2018) . European Language Resources Association, pp. 3890-3894 . < https://arxiv.org/pdf/1704.05347.pdf >
The recent years have seen a revival of interest in textual entailment, sparked by i) the emergence of powerful deep neural network learners for natural language processing and ii) the timely development of large-scale evaluation datasets such as SNL
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=arXiv_dedup_::61005b960a292f262965a6994897f602
https://pure.itu.dk/portal/da/publications/00ef2199-9e6d-4a52-ab71-0e2a1256cfdf
https://pure.itu.dk/portal/da/publications/00ef2199-9e6d-4a52-ab71-0e2a1256cfdf