Recent Submissions

  • Padding Aware Neurons 

    Garcia Gasulla, Dario; Giménez Ábalos, Víctor; Martin Torres, Pablo Agustin (Institute of Electrical and Electronics Engineers (IEEE), 2023)
    Conference report
    Open Access
    Convolutional layers are a fundamental component of most image-related models. These layers often implement by default a static padding policy (e.g. zero padding), to control the scale of the internal representations, and ...
  • Analyzing text representations by measuring task alignment 

    González Gutiérrez, César; Primadhanty, Audi; Cazzaro, Francesco; Quattoni, Ariadna Julieta (Association for Computational Linguistics, 2023)
    Conference report
    Open Access
    Textual representations based on pre-trained language models are key, especially in few-shot learning scenarios. What makes a representation good for text classification? Is it due to the geometric properties of the space ...
  • Applying generative models and transfer learning to physiological data classification 

    Núñez Rodríguez, José Fernando; Arjona Martínez, Jamie; Tormos Llorente, Adrián; Garcia Gasulla, Dario; Béjar Alonso, Javier (IOS Press, 2023-11-01)
    Conference report
    Open Access
    The scarcity and imbalance of datasets for training deep learning models in a specific task is a common problem. This is especially true in the physiological domain where many applications use complex data collection ...
  • Brealing through the traffic congestion: asynchronous time series data integration and XGBoost for accurate traffic density prediction 

    Garcia Climent, Eloi; Serrat Piè, Carles; Xhafa Xhafa, Fatos (Institute of Electrical and Electronics Engineers (IEEE), 2023)
    Conference report
    Open Access
  • Measuring the mixing of contextual information in the transformer 

    Ferrando Monsonís, Javier; Gallego Olsina, Gerard Ion; Ruiz Costa-Jussà, Marta (Association for Computational Linguistics, 2022)
    Conference lecture
    Open Access
    The Transformer architecture aggregates input information through the self-attention mechanism, but there is no clear understanding of how this information is mixed across the entire model. Additionally, recent works have ...
  • Explaining how transformers use context to build predictions 

    Ferrando Monsonís, Javier; Gallego Olsina, Gerard Ion; Tsiamas, Ioannis; Ruiz Costa-jussà, Marta (Association for Computational Linguistics, 2023)
    Conference lecture
    Open Access
    Language Generation Models produce words based on the previous context. Although existing methods offer input attributions as explanations for a model’s prediction, it is still unclear how prior words affect the model’s ...
  • Ensemble model-based method for time series sensors’ data validation and imputation applied to a real waste water treatment plant 

    Pascual Pañach, Josep; Sànchez-Marrè, Miquel; Cugueró Escofet, Miquel Àngel (International Environmental Modelling and Software Society (iEMSS), 2022)
    Conference lecture
    Open Access
    Intelligent Decision Support Systems (IDSSs) integrate different Artificial Intelligence (AI) techniques with the aim of taking or supporting human-like decisions. To this end, these techniques are based on the available ...
  • Recognition of conformational states of a G protein coupled receptor from molecular dynamic simulations using sampling techniques 

    Gutiérrez Mondragón, Mario Alberto; König, Caroline; Vellido Alcacena, Alfredo (Springer, 2023)
    Conference report
    Restricted access - publisher's policy
    Protein structures are complex and dynamic entities relevant to many biological processes. G-protein-coupled receptors in particular are a functionally relevant family of cell membrane proteins of interest as targets in ...
  • Translate first reorder later: leveraging monotonicity in semantic parsing 

    Cazzaro, Francesco; Locatelli, Davide; Quattoni, Ariadna Julieta; Carreras Pérez, Xavier (Association for Computational Linguistics, 2023)
    Conference report
    Open Access
    Prior work in semantic parsing has shown that conventional seq2seq models fail at compositional generalization tasks. This limitation led to a resurgence of methods that model alignments between sentences and their ...
  • Adding preferences and moral values in an agent-based simulation framework for high-performance computing 

    Marin Gutierrez, David; Vázquez Salceda, Javier; Álvarez Napagao, Sergio; Gnatyshak, Dmitry (2023)
    Conference report
    Open Access
    Agent-Based Simulation is a suitable approach used now-a-days to simulate and analyze complex societal environments and scenarios. Current Agent-Based Simulation frameworks either scale quite well in computation but implement ...
  • Explainable agents adapt to human behaviour 

    Tormos Llorente, Adrián; Giménez Ábalos, Víctor; Domènech Vila, Marc; Gnatyshak, Dmitry; Álvarez Napagao, Sergio; Vázquez Salceda, Javier (2023)
    Conference report
    Open Access
    When integrating artificial agents into physical or digital environments that are shared with humans, agents are often equipped with opaque Machine Learning methods to enable adapting their behaviour to dynamic human needs ...
  • Analyzing multiple conflicts in SAT: an experimental evaluation 

    Oliveras Llunell, Albert; Rodríguez Carbonell, Enric; Zhao, Rui (EasyChair Publications, 2023)
    Conference report
    Open Access
    Unit propagation and conflict analysis are two essential ingredients of CDCL SAT Solving. The order in which unit propagation is computed does not matter when no conflict is found, because it is well known that there exists ...

View more