no code implementations • EAMT 2020 • Miguel Domingo, Mercedes García-Martínez, Álvaro Peris, Alexandre Helle, Amando Estela, Laurent Bié, Francisco Casacuberta, Manuel Herranz
Adaptive neural machine translation systems, able to incrementally update the underlying models under an online learning regime, have been proven to be useful to improve the efficiency of this workflow.
no code implementations • IWSLT (EMNLP) 2018 • Miguel Domingo, Francisco Casacuberta
This makes historical documents to be hard to comprehend by contemporary people and, thus, limits their accessibility to scholars specialized in the time period in which a certain document was written.
no code implementations • AMTA 2022 • Salvador Carrión-Ponz None, Francisco Casacuberta
First, we describe the catastrophic forgetting phenomenon as a function of the number of tasks learned (language pairs) and the ratios of past data used during the learning of the new task.
no code implementations • AMTA 2022 • Martín Quesada Zaragoza, Francisco Casacuberta
Cross-lingual alignment methods for monolingual language representations have received notable attention in recent years.
no code implementations • AMTA 2022 • Salvador Carrión-Ponz, Francisco Casacuberta
Next, we study the ability of these models to mitigate the effects of catastrophic forgetting in machine translation.
no code implementations • EAMT 2020 • Daniel Marín Buj, Daniel Ibáñez García, Zuzanna Parcheta, Francisco Casacuberta
In this paper, we present a machine translation system implemented by the Translation Centre for the Bodies of the European Union (CdT).
no code implementations • MTSummit 2021 • Ángel Navarro, Francisco Casacuberta
The quality of the translations generated by Machine Translation (MT) systems has highly improved through the years and but we are still far away to obtain fully automatic high-quality translations.
no code implementations • 9 Jul 2024 • Angel Navarro, Francisco Casacuberta
Pre-trained large language models (LLM) are starting to be widely used in many applications.
no code implementations • 9 Feb 2023 • Salvador Carrión, Francisco Casacuberta
We present AutoNMT, a framework to streamline the research of seq-to-seq models by automating the data pipeline (i. e., file management, data preprocessing, and exploratory analysis), automating experimentation in a toolkit-agnostic manner, which allows users to use either their own models or existing seq-to-seq toolkits such as Fairseq or OpenNMT, and finally, automating the report generation (plots and summaries).
no code implementations • 14 Nov 2022 • Francisco Casacuberta, Alexandru Ceausu, Khalid Choukri, Miltos Deligiannis, Miguel Domingo, Mercedes García-Martínez, Manuel Herranz, Guillaume Jacquet, Vassilis Papavassiliou, Stelios Piperidis, Prokopis Prokopidis, Dimitris Roussis, Marwa Hadj Salah
This work presents the results of the machine translation (MT) task from the Covid-19 MLIA @ Eval initiative, a community effort to improve the generation of MT systems focused on the current Covid-19 crisis.
1 code implementation • 2 Feb 2021 • Miguel Domingo, Francisco Casacuberta
Once the user is satisfied with the system's hypothesis and validates it, the system adapts its model following an online learning strategy.
no code implementations • 8 Oct 2019 • Miguel Domingo, Francisco Casacuberta
Modernization aims at breaking this language barrier by generating a new version of a historical document, written in the modern version of the document's original language.
no code implementations • WS 2019 • Zuzanna Parcheta, Germ{\'a}n Sanchis-Trilles, Francisco Casacuberta
The filtering task of noisy parallel corpora in WMT2019 aims to challenge participants to create filtering methods to be useful for training machine translation systems.
no code implementations • 1 Jul 2019 • Miguel Domingo, Francisco Casacuberta
This is due to the language barrier inherent in human language and the linguistic properties of these documents.
no code implementations • WS 2019 • Miguel Domingo, Mercedes García-Martínez, Álvaro Peris, Alexandre Helle, Amando Estela, Laurent Bié, Francisco Casacuberta, Manuel Herranz
A common use of machine translation in the industry is providing initial translation hypotheses, which are later supervised and post-edited by a human expert.
1 code implementation • ACL 2019 • Miguel Domingo, Mercedes García-Martínez, Amando Estela, Laurent Bié, Alexandre Helle, Álvaro Peris, Francisco Casacuberta, Manuerl Herranz
We introduce a demonstration of our system, which implements online learning for neural machine translation in a production environment.
no code implementations • 30 May 2019 • Álvaro Peris, Francisco Casacuberta
We show that, following this framework, we approximately halve the effort spent for correcting the outputs generated by the automatic systems.
1 code implementation • ACL 2019 • Álvaro Peris, Francisco Casacuberta
We present a demonstration of a neural interactive-predictive system for tackling multimodal sequence to sequence tasks.
no code implementations • 20 Dec 2018 • Miguel Domingo, Mercedes Garcıa-Martınez, Alexandre Helle, Francisco Casacuberta, Manuel Herranz
Separating punctuation and splitting tokens into words or subwords has proven to be helpful to reduce vocabulary and increase the number of examples of each word, improving the translation quality.
1 code implementation • CONLL 2018 • Álvaro Peris, Francisco Casacuberta
We study the application of active learning techniques to the translation of unbounded data streams via interactive neural machine translation.
1 code implementation • 9 Jul 2018 • Álvaro Peris, Francisco Casacuberta
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.
1 code implementation • 10 Feb 2018 • Álvaro Peris, Francisco Casacuberta
We show that a neural machine translation system can be rapidly adapted to a specific domain, exclusively by means of online learning techniques.
1 code implementation • 10 Jun 2017 • Álvaro Peris, Luis Cebrián, Francisco Casacuberta
Neural machine translation has meant a revolution of the field.
1 code implementation • 7 Apr 2017 • Marc Bolaños, Álvaro Peris, Francisco Casacuberta, Sergi Soler, Petia Radeva
We propose a novel methodology that exploits information from temporally neighboring events, matching precisely the nature of egocentric sequences.
1 code implementation • 16 Dec 2016 • Álvaro Peris, Mara Chinea-Rios, Francisco Casacuberta
We address the data selection problem in statistical machine translation (SMT) as a classification task.
1 code implementation • 12 Dec 2016 • Marc Bolaños, Álvaro Peris, Francisco Casacuberta, Petia Radeva
In this paper, we address the problem of visual question answering by proposing a novel model, called VIBIKNet.
1 code implementation • 12 Apr 2016 • Álvaro Peris, Marc Bolaños, Petia Radeva, Francisco Casacuberta
Although traditionally used in the machine translation field, the encoder-decoder framework has been recently applied for the generation of video and image descriptions.
no code implementations • LREC 2014 • Mara Chinea Rios, Germ{\'a}n Sanchis-Trilles, Daniel Ortiz-Mart{\'\i}nez, Francisco Casacuberta
Whenever the quality provided by a machine translation system is not enough, a human expert is required to correct the sentences provided by the machine translation system.
no code implementations • LREC 2014 • Nancy Underwood, Bartolom{\'e} Mesa-Lao, Mercedes Garc{\'\i}a Mart{\'\i}nez, Michael Carl, Vicent Alabau, Jes{\'u}s Gonz{\'a}lez-Rubio, Luis A. Leiva, Germ{\'a}n Sanchis-Trilles, Daniel Ort{\'\i}z-Mart{\'\i}nez, Francisco Casacuberta
This paper describes the field trial and subsequent evaluation of a post-editing workbench which is currently under development in the EU-funded CasMaCat project.
no code implementations • EACL 2014 • Vicent Alabau, Christian Buck, Michael Carl, Francisco Casacuberta, Mercedes García-Martínez, Ulrich Germann, Jesús González-Rubio, Robin Hill, Philipp Koehn, Luis Leiva, Bartolomé Mesa-Lao, Daniel Ortiz-Martínez, Herve Saint-Amand, Germán Sanchis Trilles, Chara Tsoukala