Zobrazeno 1 - 10
of 19
pro vyhledávání: '"Lina Maria Rojas-Barahona"'
Publikováno v:
XXXIVe Journées d'Études sur la Parole -- JEP 2022.
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::dd37326118d8f7046012c14a58a583d2
https://doi.org/10.21437/jep.2022-53
https://doi.org/10.21437/jep.2022-53
Publikováno v:
ACL/IJCNLP (Findings)
Knowledge Graph (KG) completion has been excessively studied with a massive number of models proposed for the Link Prediction (LP) task. The main limitation of such models is their insensitivity to time. Indeed, the temporal aspect of stored facts is
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::fcdf01bac09f8c2cce82f1d905bb99ca
https://doi.org/10.18653/v1/2021.findings-acl.292
https://doi.org/10.18653/v1/2021.findings-acl.292
Autor:
Lina Maria Rojas-Barahona
Publikováno v:
Language and Linguistics Compass. 10:701-719
Research and industry are becoming more and more interested in finding automatically the polarised opinion of the general public regarding a specific subject. The advent of social networks has opened the possibility of having access to massive blogs,
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::38086e2abc41ff8db180fa4fe8b851be
https://doi.org/10.1111/lnc3.12228
https://doi.org/10.1111/lnc3.12228
Autor:
Milica Gasic, Paweł Budzianowski, Steve Young, Stefan Ultes, Iñigo Casanueva, Yen-Chen Wu, Bo-Hsiang Tseng, Lina Maria Rojas-Barahona
Publikováno v:
SIGDIAL Conference
Statistical spoken dialogue systems usually rely on a single- or multi-domain dialogue model that is restricted in its capabilities of modelling complex dialogue structures, e.g., relations. In this work, we propose a novel dialogue model that is cen
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::c69b40e28a40deafb6cbc174c63d39ad
http://arxiv.org/abs/1901.01466
http://arxiv.org/abs/1901.01466
Autor:
Bo-Hsiang Tseng, Milica Gasic, Clare Mansfield, Stefan Ultes, Lina Maria Rojas-Barahona, Osman Ramadan, Yinpei Dai, Michael Crawford
Publikováno v:
Louhi@EMNLP
In recent years, we have seen deep learning and distributed representations of words and sentences make impact on a number of natural language processing tasks, such as similarity, entailment and sentiment analysis. Here we introduce a new task: unde
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::58e3f457cda2042f6cb0867fd46603cc
http://arxiv.org/abs/1809.00640
http://arxiv.org/abs/1809.00640
Autor:
Kyusong Lee, Tiancheng Zhao, Stefan Ultes, Lina Maria Rojas-Barahona, Eli Pincus, Maxine Eskenazi, David Traum
Publikováno v:
Lecture Notes in Electrical Engineering ISBN: 9783319921075
IWSDS
IWSDS
Collecting a large amount of real human-computer interaction data in various domains is a cornerstone in the development of better data-driven spoken dialog systems. The DialPort project is creating a portal to collect a constant stream of real user
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::2cb6084eec194f0f30b980ee435d30c3
https://doi.org/10.1007/978-3-319-92108-2_10
https://doi.org/10.1007/978-3-319-92108-2_10
Autor:
Lina Maria Rojas-Barahona, Milica Gasic, Stefan Ultes, Bo-Hsiang Tseng, Iñigo Casanueva, Paweł Budzianowski, Pei-Hao Su
Publikováno v:
NAACL-HLT (2)
Reinforcement learning (RL) is a promising approach to solve dialogue policy optimisation. Traditional RL algorithms, however, fail to scale to large domains due to the curse of dimensionality. We propose a novel Dialogue Management architecture, bas
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::2a647a69972f615afe19346b273c2c47
Autor:
Stefan Ultes, Paweł Budzianowski, Iñigo Casanueva, Steve Young, Nikola Mrkšić, Tsung-Hsien Wen, Milica Gasic, Lina Maria Rojas-Barahona, Pei-Hao Su
Publikováno v:
INTERSPEECH
Copyright © 2017 ISCA. Learning suitable and well-performing dialogue behaviour in statistical spoken dialogue systems has been in the focus of research for many years. While most work which is based on reinforcement learning employs an objective me
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::c47739fcb2f3ad0bf6afa33726622a4e
https://doi.org/10.21437/interspeech.2017-1032
https://doi.org/10.21437/interspeech.2017-1032
Autor:
Steve Young, Pei-Hao Su, Nikola Mrkšić, Paweł Budzianowski, Dongho Kim, Stefan Ultes, Iñigo Casanueva, David Vandyke, Milica Gasic, Tsung-Hsien Wen, Lina Maria Rojas-Barahona
Publikováno v:
ACL (System Demonstrations)
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::a8aac1165836e02d86d0c74e55abf82b
https://doi.org/10.18653/v1/p17-4013
https://doi.org/10.18653/v1/p17-4013
Autor:
Steve Young, Paweł Budzianowski, Lina Maria Rojas-Barahona, Tsung-Hsien Wen, Stefan Ultes, Iñigo Casanueva, Milica Gasic, Nikola Mrkšić, Pei-Hao Su
Publikováno v:
SIGDIAL Conference
Reinforcement learning is widely used for dialogue policy optimization where the reward function often consists of more than one component, e.g., the dialogue success and the dialogue length. In this work, we propose a structured method for finding a
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::0a2664c88939695a700e6194ee7d5168
https://doi.org/10.18653/v1/w17-5509
https://doi.org/10.18653/v1/w17-5509