Top-down string-to-dependency Neural Machine Translation
This addresses translation accuracy for long sentences in NLP, but it is incremental as it builds on existing syntax-incorporation approaches.
The paper tackles the problem of neural machine translation struggling with long, rare inputs by proposing a top-down syntactic decoder that generates dependency trees, showing it generalizes better than conventional methods on unseen long inputs.
Most of modern neural machine translation (NMT) models are based on an encoder-decoder framework with an attention mechanism. While they perform well on standard datasets, they can have trouble in translation of long inputs that are rare or unseen during training. Incorporating target syntax is one approach to dealing with such length-related problems. We propose a novel syntactic decoder that generates a target-language dependency tree in a top-down, left-to-right order. Experiments show that the proposed top-down string-to-tree decoding generalizes better than conventional sequence-to-sequence decoding in translating long inputs that are not observed in the training data.