《Neural Machine Translation》.pdf

Neural Machine Translation by Jointly Learning to Align and Translate Dzmitry Bahdanau KyungHyun Cho Yoshua Bengio ´ ´ ´ ´ Jacobs University, Germany Universite de Montreal Universite de Montreal CIFAR Senior Fellow 4 1 0 2 Abstract p e Neural machine translation is a recently proposed approach to machine transla- S tion. Unlike the traditional statistical machine translation, the neural machine 4 translation aims at building a single neural network that can be jointly tuned to maximize the translation performance. The models proposed recently for neural ] machine translation often belong to a family of encoder–decoders and encodes L a source sentence into a fixed-length vector from which a decoder generates a C translation. In this paper, we conjecture that the use of a fixed-length vector is a s. bottleneck in improving the performance of this basic encoder–decoder architec- c ture, and propose to extend this by allowing a model to automatically (soft-)search [ for parts of a source sentence that are relevant to predicting a target word, without 2 having to form these parts as a hard segment explicitly. With this new approac