S
Sergio Barrachina
Researcher at James I University
Publications - 50
Citations - 984
Sergio Barrachina is an academic researcher from James I University. The author has contributed to research in topics: Computer science & Parallel algorithm. The author has an hindex of 13, co-authored 44 publications receiving 897 citations.
Papers
More filters
Journal ArticleDOI
Statistical approaches to computer-assisted translation
Sergio Barrachina,Oliver Bender,Francisco Casacuberta,Jorge Civera,Elsa Cubel,Shahram Khadivi,Antonio Lagarda,Hermann Ney,Jesús Tomás,Enrique Vidal,Juan-Miguel Vilar +10 more
TL;DR: Alignment templates, phrase-based models, and stochastic finite-state transducers are used to develop computer-assisted translation systems in a European project in two real tasks.
Proceedings ArticleDOI
Evaluation and tuning of the Level 3 CUBLAS for graphics processors
TL;DR: This paper evaluates the performance of the Level 3 operations in CUBLAS, the implementation of BIAS for NVIDIAreg GPUs with unified architecture, and gains insights on the quality of the kernels in the library.
Journal ArticleDOI
Some approaches to statistical and finite-state speech-to-speech translation
Francisco Casacuberta,Hermann Ney,Franz Josef Och,Enrique Vidal,Juan Miguel Vilar,Sergio Barrachina,Ismael García-Varea,David Llorens,César E. Martínez,S. Molau,Francisco Nevado,Moisés Pastor,David Picó,Alberto Sanchis,Christoph Tillmann +14 more
TL;DR: Speech-input translation can be properly approached as a pattern recognition problem by means of statistical alignment models and stochastic finite-state transducers, and some specific models are presented.
Book ChapterDOI
Solving Dense Linear Systems on Graphics Processors
TL;DR: Several algorithms to compute the solution of a linear system of equations on a GPU, as well as general techniques to improve their performance, such as padding and hybrid GPU-CPU computation are presented.
Journal IssueDOI
Exploiting the capabilities of modern GPUs for dense matrix computations
Sergio Barrachina,Maribel Castillo,Francisco D. Igual,Rafael Mayo,Enrique S. Quintana-Ortí,Gregorio Quintana-Ortí +5 more
TL;DR: This work compares single and double precision performance of a modern GPU with unified architecture, and shows how iterative refinement with mixed precision can be used to regain full accuracy in the solution of linear systems.