用自顶向下生成依存树提升长句翻译效果
Top-down string-to-dependency Neural Machine Translation
- 自顶向下逐节点构建目标语言依存树
- 长句子翻译准确率显著优于传统序列模型
- 适合处理训练中未见的长难句场景
当前主流神经机器翻译模型基于编码器-解码器框架与注意力机制。尽管在标准数据集上表现良好，但在翻译长输入时若这些输入在训练中罕见或未出现，性能会下降。引入目标语言语法结构是缓解此类问题的方法之一。本文提出一种新型句法解码器，以自顶向下、从左到右的顺序生成目标语言的依存树。实验表明，该自顶向下字符串到依存树的解码方式，在处理训练数据中未观察到的长输入时，泛化能力优于传统的序列到序列解码。
原文摘要 · Abstract (English)
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.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。