CNN Is All You Need
This addresses the performance bottleneck in sequence-to-sequence learning for machine translation by improving CNN-based models, though it is incremental as it builds on existing CNN approaches.
The paper tackled the lack of history sensitivity in CNNs for sequence-to-sequence learning by introducing PoseNet, an extended CNN model with enhanced position-sensitivity, achieving BLEU scores of 33-36 on English-to-German and 44-46 on English-to-French translation tasks.
The Convolution Neural Network (CNN) has demonstrated the unique advantage in audio, image and text learning; recently it has also challenged Recurrent Neural Networks (RNNs) with long short-term memory cells (LSTM) in sequence-to-sequence learning, since the computations involved in CNN are easily parallelizable whereas those involved in RNN are mostly sequential, leading to a performance bottleneck. However, unlike RNN, the native CNN lacks the history sensitivity required for sequence transformation; therefore enhancing the sequential order awareness, or position-sensitivity, becomes the key to make CNN the general deep learning model. In this work we introduce an extended CNN model with strengthen position-sensitivity, called PoseNet. A notable feature of PoseNet is the asymmetric treatment of position information in the encoder and the decoder. Experiments shows that PoseNet allows us to improve the accuracy of CNN based sequence-to-sequence learning significantly, achieving around 33-36 BLEU scores on the WMT 2014 English-to-German translation task, and around 44-46 BLEU scores on the English-to-French translation task.