Attention is All You Need

This paper introduces the Transformer, a novel neural network architecture based solely on attention mechanisms, dispensing with recurrence and convolutions entirely. It demonstrates that this architecture can be trained faster and achieves state-of-the-art results in machine translation tasks. The core idea is to use self-attention to relate different positions of the input sequence in order to compute a representation of the entire sequence. This approach significantly improves parallelization and requires less time to train while achieving better performance compared to previous recurrent and convolutional models.