Zobrazeno 1 - 10
of 61
pro vyhledávání: '"Carreras, Xavier"'
Recent advances in deep learning models for sequence classification have greatly improved their classification accuracy, specially when large training sets are available. However, several works have suggested that under some settings the predictions
Externí odkaz:
http://arxiv.org/abs/2210.13082
Prior work in semantic parsing has shown that conventional seq2seq models fail at compositional generalization tasks. This limitation led to a resurgence of methods that model alignments between sentences and their corresponding meaning representatio
Externí odkaz:
http://arxiv.org/abs/2210.04878
Autor:
Ballesteros, Miguel, Carreras, Xavier
We present a neural transition-based parser for spinal trees, a dependency representation of constituent trees. The parser uses Stack-LSTMs that compose constituent nodes with dependency-based derivations. In experiments, we show that this model adap
Externí odkaz:
http://arxiv.org/abs/1709.00489
We present a solution to scale spectral algorithms for learning sequence functions. We are interested in the case where these functions are sparse (that is, for most sequences they return 0). Spectral algorithms reduce the learning problem to the tas
Externí odkaz:
http://arxiv.org/abs/1706.02857
We investigate the problem of inducing word embeddings that are tailored for a particular bilexical relation. Our learning algorithm takes an existing lexical vector space and compresses it such that the resulting word embeddings are good predictors
Externí odkaz:
http://arxiv.org/abs/1412.7004
This paper re-visits the spectral method for learning latent variable models defined in terms of observable operators. We give a new perspective on the method, showing that operators can be recovered by minimizing a loss defined on a finite subset of
Externí odkaz:
http://arxiv.org/abs/1206.6393
Autor:
Carreras, Xavier, Marquez, Lluis
Publikováno v:
Proceedings of RANLP-2001, pp. 58-64, Bulgaria, 2001
This paper describes a set of comparative experiments for the problem of automatically filtering unwanted electronic mail messages. Several variants of the AdaBoost algorithm with confidence-rated predictions [Schapire & Singer, 99] have been applied
Externí odkaz:
http://arxiv.org/abs/cs/0109015
Publikováno v:
Digital.CSIC. Repositorio Institucional del CSIC
instname
Ribeiro, J, Narayan, S, Cohen, S & Carreras, X 2018, Local String Transduction as Sequence Labeling . in 27th International Conference on Computational Linguistics (COLING 2018) . Santa Fe, New-Mexico, USA, pp. 1360-1371, 27th International Conference on Computational Linguistics, Sante Fe, New Mexico, United States, 20/08/18 . < http://coling2018.org/wp-content/uploads/2018/08/coling18-main.pdf >
Ribeiro, J, Narayan, S, Cohen, S & Carreras, X 2018, Local String Transduction as Sequence Labeling . in 27th International Conference on Computational Linguistics (COLING 2018) . Santa Fe, New-Mexico, USA, pp. 1360-1371, 27th International Conference on Computational Linguistics, Sante Fe, United States, 20/08/18 .
instname
Ribeiro, J, Narayan, S, Cohen, S & Carreras, X 2018, Local String Transduction as Sequence Labeling . in 27th International Conference on Computational Linguistics (COLING 2018) . Santa Fe, New-Mexico, USA, pp. 1360-1371, 27th International Conference on Computational Linguistics, Sante Fe, New Mexico, United States, 20/08/18 . < http://coling2018.org/wp-content/uploads/2018/08/coling18-main.pdf >
Ribeiro, J, Narayan, S, Cohen, S & Carreras, X 2018, Local String Transduction as Sequence Labeling . in 27th International Conference on Computational Linguistics (COLING 2018) . Santa Fe, New-Mexico, USA, pp. 1360-1371, 27th International Conference on Computational Linguistics, Sante Fe, United States, 20/08/18 .
[EN]We show that the general problem of string transduction can be reduced to the problem of sequence labeling. While character deletion and insertions are allowed in string transduction, they do not exist in sequence labeling. We show how to overcom
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::cbe8dd8a45d261b3be65730f0423fdca
Autor:
Nicolás, David, Camós-Carreras, Anna, Spencer, Felipe, Arenas, Andrea, Butori, Eugenia, Maymó, Pol, Anmella, Gerard, Torrallardona-Murphy, Orla, Alves, Eduarda, García, Laura, Pereta, Irene, Castells, Eva, Seijas, Nuria, Ibáñez, Begoña, Grané, Carme, Bodro, Marta, Cardozo, Celia, Barroso, Sonia, Olive, Victoria, Tortajada, Marta, Hernández, Carme, Cucchiari, David, Coloma, Emmanuel, Pericàs, Juan M, Martinez, Gemma, Castells, Antoni, Feu, Faust, Cadenas, Roser, Varela, Pilar, Pericás, Juan M, Calvo, Júlia, López-Soto, Alfons, Soriano, Álex, Nicolás, Josep M, Llufriu, Sara, Camos, Anna, Escudero, Lucía, Dotti, Marina, Teresa Carrión, M, Opazo, Valeria, Parrado, Alba, Giralt, Joan, Bernal, Carolina, Romero, Barbara, Boquera, Clàudia, Sánchez, Miriam, Feu, Silvia, Casablanca, Anna, Cayado, Ana, Carreras, Xavier, Pablo Figueroa, J, Marín, Sara, Castro, Rafa, Oliva, Cristian, Torrallardona, Orla, Maymo, Pol, Serralabós, Jùlia, Alvès, Elisenda, Rabaneda, Neus, Hidalgo, Judit, Avalos, Maribel, Carbonell, Anna, Subirana, Núria, Navas, Regina, Aranda, Carmen, Rodríguez, Magali, Salas, Marta, Suárez, Adolfo, Fernández, Ana, Martínez, Alba, Barta, Ariadna, Escobar, Cristina, Moreno, Laura, Jawara, Mohammed, Cano, Susana, Román, Mariana, Martinez, Maria, Jiménez, David, Rosero, Elisabeth, Llop, Lourdes, Asenjo, Maria
Publikováno v:
Dipòsit Digital de la UB
Universidad de Barcelona
Open Forum Infectious Diseases
Scientia
Universidad de Barcelona
Open Forum Infectious Diseases
Scientia
Background During the coronavirus disease 2019 (COVID-19) outbreaks, health care workers (HCWs) are at a high risk of infection. Strategies to reduce in-hospital transmission between HCWs and to safely manage infected HCWs are lacking. Our aim was to
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::266d44c4209bf893035d6d7e56220226
http://hdl.handle.net/2445/184221
http://hdl.handle.net/2445/184221
Akademický článek
Tento výsledek nelze pro nepřihlášené uživatele zobrazit.
K zobrazení výsledku je třeba se přihlásit.
K zobrazení výsledku je třeba se přihlásit.