Zobrazeno 1 - 10
of 1 710
pro vyhledávání: '"Utiyama AS"'
Autor:
Wang, Yiran, Utiyama, Masao
Unsupervised parsing, also known as grammar induction, aims to infer syntactic structure from raw text. Recently, binary representation has exhibited remarkable information-preserving capabilities at both lexicon and syntax levels. In this paper, we
Externí odkaz:
http://arxiv.org/abs/2410.04074
Autor:
Wang, Yiran, Utiyama, Masao
Recently, binary representation has been proposed as a novel representation that lies between continuous and discrete representations. It exhibits considerable information-preserving capability when being used to replace continuous input vectors. In
Externí odkaz:
http://arxiv.org/abs/2406.07812
Autor:
Deguchi, Hiroyuki, Sakai, Yusuke, Kamigaito, Hidetaka, Watanabe, Taro, Tanaka, Hideki, Utiyama, Masao
Minimum Bayes risk (MBR) decoding achieved state-of-the-art translation performance by using COMET, a neural metric that has a high correlation with human evaluation. However, MBR decoding requires quadratic time since it computes the expected score
Externí odkaz:
http://arxiv.org/abs/2402.11197
Autor:
Zhang, Zhuosheng, Chen, Kehai, Wang, Rui, Utiyama, Masao, Sumita, Eiichiro, Li, Zuchao, Zhao, Hai
Representation learning is the foundation of natural language processing (NLP). This work presents new methods to employ visual information as assistant signals to general NLP tasks. For each sentence, we first retrieve a flexible number of images ei
Externí odkaz:
http://arxiv.org/abs/2301.03344
Discriminative pre-trained language models (PLMs) learn to predict original texts from intentionally corrupted ones. Taking the former text as positive and the latter as negative samples, the PLM can be trained effectively for contextualized represen
Externí odkaz:
http://arxiv.org/abs/2212.00460
Autor:
CARLOS AUGUSTO METIDIERI MENEGOZZO, CRISTIANE PRESTES DA ROSA OLIVEIRA, ROBERTO RASSLAN, FELIPE KFOURI, ALBERTO BITRAN, RENATO SILVEIRA LEAL, SERGIO HENRIQUE BASTOS DAMOUS, EDIVALDO MASSAZO UTIYAMA
Publikováno v:
Revista do Colégio Brasileiro de Cirurgiões, Vol 51 (2024)
ABSTRACT The role of wound protectors in laparoscopic surgeries is highly controversial in the literature. Some studies demonstrate their benefit in reducing the rate of surgical site infections; however, these results are not reproducible across all
Externí odkaz:
https://doaj.org/article/9f0cf75a65c64aaf8d935877ee246721
Autor:
Estevão Bassi, Bruno Martins Tomazini, Bárbara Vieira Carneiro, Amanda Rodrigues de Oliveira Siqueira, Sara Rodrigues de Oliveira Siqueira, Thais Guimarães, Fernando da Costa Ferreira Novo, Edivaldo Massazo Utiyama, Paolo Pelosi, Luiz Marcelo Sá Malbouisson
Publikováno v:
Brazilian Journal of Anesthesiology, Vol 74, Iss 3, Pp 744431- (2024)
Background: Systemic inflammatory responses mimicking infectious complications are often present in surgical patients. Methods: The objective was to assess the association between withholding early antimicrobial therapy while investigating alternativ
Externí odkaz:
https://doaj.org/article/b2332c82e89a4aa39d283f2ff7702134
Autor:
Estevão Bassi, Camila Trevizani Merighi, Carlos Issamu Tomizuka, Thais Guimarães, Fernando da Costa Ferreira Novo, Sergio Henrique Bastos Damous, Edivaldo Massazo Utiyama, Luiz Marcelo Sá Malbouisson
Publikováno v:
Brazilian Journal of Anesthesiology, Vol 74, Iss 3, Pp 744454- (2024)
Background: Pneumonia occurs in about 20% of trauma patients with pulmonary contusions. This study aims to evaluate the association between empirical antibiotic therapy and nosocomial pneumonia in this population. Methods: Retrospective cohort of adu
Externí odkaz:
https://doaj.org/article/fe9668844c4c48b997fcff62dca0d82b
Conversational machine reading (CMR) requires machines to communicate with humans through multi-turn interactions between two salient dialogue states of decision making and question generation processes. In open CMR settings, as the more realistic sc
Externí odkaz:
http://arxiv.org/abs/2108.12599
Autor:
Li, Zuchao, Parnow, Kevin, Zhao, Hai, Zhang, Zhuosheng, Wang, Rui, Utiyama, Masao, Sumita, Eiichiro
Though the pre-trained contextualized language model (PrLM) has made a significant impact on NLP, training PrLMs in languages other than English can be impractical for two reasons: other languages often lack corpora sufficient for training powerful P
Externí odkaz:
http://arxiv.org/abs/2107.12627