Álvaro Peris
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Álvaro Peris has a total of 11 co-authors. Here are the top co-authors by number of shared publications:
- Francisco Casacuberta
Universitat Politècnica de València
9 shared publications Francisco Casacuberta
6 shared publications- Marc Bolaños
Universitat de Barcelona
4 shared publications - Petia Radeva
Universitat de Barcelona
4 shared publications - Miguel Domingo
Universitat Politècnica de València
2 shared publications Mara Chinea-Rios
1 shared publication
Latest Publications
- 1 November 2019
- Published by Elsevier in Computer Speech & Language
- Vol. 58, 98-126
Online learning for effort reduction in interactive neural machine translation
- 1 November 2019
- Published by Elsevier in Computer Speech & Language
- Vol. 58, 98-126
Online learning for effort reduction in interactive neural machine translation
- 22 September 2019
- Published by Springer Nature
- p. 16-28
Interactive-Predictive Neural Multimodal Systems
- 22 September 2019
- Published by Springer Nature
- p. 16-28
Interactive-Predictive Neural Multimodal Systems
- Open Access
Online Learning for Effort Reduction in Interactive Neural Machine Translation
Neural machine translation systems require large amounts of training data and resources. Even with this, the quality of the translations may be insufficient for some users or domains. In such cases, the output of the system must be revised by a human agent. This can be done in a post-editing stage or following an interactive machine translation protocol. We explore the incremental update of neural machine translation systems during the post-editing or interactive translation processes. Such modifications aim to incorporate the new knowledge, from the edited sentences, into the translation system. Updates to the model are performed on-the-fly, as sentences are corrected, via online learning techniques. In addition, we implement a novel interactive, adaptive system, able to react to single-character interactions. This system greatly reduces the human effort required for obtaining high-quality translations. In order to stress our proposals, we conduct exhaustive experiments varying the amount and type of data available for training. Results show that online learning effectively achieves the objective of reducing the human effort required during the post-editing or the interactive machine translation stages. Moreover, these adaptive systems also perform well in scenarios with scarce resources. We show that a neural machine translation system can be rapidly adapted to a specific domain, exclusively by means of online learning techniques.
Open AccessOnline Learning for Effort Reduction in Interactive Neural Machine Translation
Neural machine translation systems require large amounts of training data and resources. Even with this, the quality of the translations may be insufficient for some users or domains. In such cases, the output of the system must be revised by a human agent. This can be done in a post-editing stage or following an interactive machine translation protocol. We explore the incremental update of neural machine translation systems during the post-editing or interactive translation processes. Such modifications aim to incorporate the new knowledge, from the edited sentences, into the translation system. Updates to the model are performed on-the-fly, as sentences are corrected, via online learning techniques. In addition, we implement a novel interactive, adaptive system, able to react to single-character interactions. This system greatly reduces the human effort required for obtaining high-quality translations. In order to stress our proposals, we conduct exhaustive experiments varying the amount and type of data available for training. Results show that online learning effectively achieves the objective of reducing the human effort required during the post-editing or the interactive machine translation stages. Moreover, these adaptive systems also perform well in scenarios with scarce resources. We show that a neural machine translation system can be rapidly adapted to a specific domain, exclusively by means of online learning techniques.
- Open Access
- 1 October 2018
- Published by Charles University in Prague, Karolinum Press in The Prague Bulletin of Mathematical Linguistics
- Vol. 111 (1) , 113-124
NMT-Keras: a Very Flexible Toolkit with a Focus on Interactive NMT and Online Learning
We present NMT-Keras, a flexible toolkit for training deep learning models, which puts a particular emphasis on the development of advanced applications of neural machine translation systems, such as interactive-predictive translation protocols and long-term adaptation of the translation system via continuous learning. NMT-Keras is based on an extended version of the popular Keras library, and it runs on Theano and TensorFlow. State-of-the-art neural machine translation models are deployed and used following the high-level framework provided by Keras. Given its high modularity and flexibility, it also has been extended to tackle different problems, such as image and video captioning, sentence classification and visual question answering.
- 1 October 2018
- Published by Charles University in Prague, Karolinum Press in The Prague Bulletin of Mathematical Linguistics
- Vol. 111 (1) , 113-124
Open AccessNMT-Keras: a Very Flexible Toolkit with a Focus on Interactive NMT and Online Learning
We present NMT-Keras, a flexible toolkit for training deep learning models, which puts a particular emphasis on the development of advanced applications of neural machine translation systems, such as interactive-predictive translation protocols and long-term adaptation of the translation system via continuous learning. NMT-Keras is based on an extended version of the popular Keras library, and it runs on Theano and TensorFlow. State-of-the-art neural machine translation models are deployed and used following the high-level framework provided by Keras. Given its high modularity and flexibility, it also has been extended to tackle different problems, such as image and video captioning, sentence classification and visual question answering.
- Open Access
NMT-Keras: a Very Flexible Toolkit with a Focus on Interactive NMT and Online Learning
We present NMT-Keras, a flexible toolkit for training deep learning models, which puts a particular emphasis on the development of advanced applications of neural machine translation systems, such as interactive-predictive translation protocols and long-term adaptation of the translation system via continuous learning. NMT-Keras is based on an extended version of the popular Keras library, and it runs on Theano and Tensorflow. State-of-the-art neural machine translation models are deployed and used following the high-level framework provided by Keras. Given its high modularity and flexibility, it also has been extended to tackle different problems, such as image and video captioning, sentence classification and visual question answering.
Open AccessNMT-Keras: a Very Flexible Toolkit with a Focus on Interactive NMT and Online Learning
We present NMT-Keras, a flexible toolkit for training deep learning models, which puts a particular emphasis on the development of advanced applications of neural machine translation systems, such as interactive-predictive translation protocols and long-term adaptation of the translation system via continuous learning. NMT-Keras is based on an extended version of the popular Keras library, and it runs on Theano and Tensorflow. State-of-the-art neural machine translation models are deployed and used following the high-level framework provided by Keras. Given its high modularity and flexibility, it also has been extended to tackle different problems, such as image and video captioning, sentence classification and visual question answering.
- Open Access
- 1 January 2018
- Published by Association for Computational Linguistics (ACL)
- p. 151-160
Active Learning for Interactive Neural Machine Translation of Data Streams
- 1 January 2018
- Published by Association for Computational Linguistics (ACL)
- p. 151-160
Open AccessActive Learning for Interactive Neural Machine Translation of Data Streams
- Open Access
- 1 January 2018
- Published by Elsevier in Journal of Visual Communication and Image Representation
- Vol. 50, 205-216
Egocentric video description based on temporally-linked sequences
- 1 January 2018
- Published by Elsevier in Journal of Visual Communication and Image Representation
- Vol. 50, 205-216
Open AccessEgocentric video description based on temporally-linked sequences
- 1 December 2017
- Published by Springer Nature in Machine Translation
- Vol. 31 (4) , 163-185
Segment-based interactive-predictive machine translation
- 1 December 2017
- Published by Springer Nature in Machine Translation
- Vol. 31 (4) , 163-185
Segment-based interactive-predictive machine translation
- Open Access
Egocentric Video Description based on Temporally-Linked Sequences
Egocentric vision consists in acquiring images along the day from a first person point-of-view using wearable cameras. The automatic analysis of this information allows to discover daily patterns for improving the quality of life of the user. A natural topic that arises in egocentric vision is storytelling, that is, how to understand and tell the story relying behind the pictures. In this paper, we tackle storytelling as an egocentric sequences description problem. We propose a novel methodology that exploits information from temporally neighboring events, matching precisely the nature of egocentric sequences. Furthermore, we present a new method for multimodal data fusion consisting on a multi-input attention recurrent network. We also publish the first dataset for egocentric image sequences description, consisting of 1,339 events with 3,991 descriptions, from 55 days acquired by 11 people. Furthermore, we prove that our proposal outperforms classical attentional encoder-decoder methods for video description.
Open AccessEgocentric Video Description based on Temporally-Linked Sequences
Egocentric vision consists in acquiring images along the day from a first person point-of-view using wearable cameras. The automatic analysis of this information allows to discover daily patterns for improving the quality of life of the user. A natural topic that arises in egocentric vision is storytelling, that is, how to understand and tell the story relying behind the pictures. In this paper, we tackle storytelling as an egocentric sequences description problem. We propose a novel methodology that exploits information from temporally neighboring events, matching precisely the nature of egocentric sequences. Furthermore, we present a new method for multimodal data fusion consisting on a multi-input attention recurrent network. We also publish the first dataset for egocentric image sequences description, consisting of 1,339 events with 3,991 descriptions, from 55 days acquired by 11 people. Furthermore, we prove that our proposal outperforms classical attentional encoder-decoder methods for video description.
- 1 September 2017
- Published by Elsevier in Computer Speech & Language
- Vol. 45, 201-220
Interactive neural machine translation
- 1 September 2017
- Published by Elsevier in Computer Speech & Language
- Vol. 45, 201-220
Interactive neural machine translation