Local Attention: Enhancing the Transformer Architecture for Efficient Time Series Forecasting | Publicación