arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1706.05565cs.CLstat.ML
收藏

基于神经短语的机器翻译

Towards Neural Phrase-based Machine Translation

Microsoft Research(微软研究院);Google(谷歌);University of Illinois at Urbana-Champaign(伊利诺伊大学厄巴纳-香槟分校);Citadel(城堡投资)

Po-Sen Huang, Chong Wang, Sitao Huang, Dengyong Zhou, Li Deng

摘要:

在本文中,我们提出了基于神经短语的机器翻译(NPMT)。我们的方法显式地使用睡眠唤醒网络(SWAN),最近提出的基于分割的序列建模方法的输出序列中的短语结构建模。为了减轻SWAN的单调对齐要求,我们引入了一个新的层来执行输入序列的(软)局部重排。与现有的神经机器翻译(NMT)方法不同,NPMT不使用基于注意力的解码机制。相反,它直接按顺序输出短语,并可以在线性时间内解码。我们的实验表明,与强大的NMT基线相比,NPMT在IWSLT 2014德语-英语/英语-德语和IWSLT 2015英语-越南语机器翻译任务上取得了优异的性能。我们还观察到,我们的方法在输出语言中产生有意义的短语。

英文摘要:

In this paper, we present Neural Phrase-based Machine Translation (NPMT). Our method explicitly models the phrase structures in output sequences using Sleep-WAke Networks (SWAN), a recently proposed segmentation-based sequence modeling method. To mitigate the monotonic alignment requirement of SWAN, we introduce a new layer to perform (soft) local reordering of input sequences. Different from existing neural machine translation (NMT) approaches, NPMT does not use attention-based decoding mechanisms. Instead, it directly outputs phrases in a sequential order and can decode in linear time. Our experiments show that NPMT achieves superior performances on IWSLT 2014 German-English/English-German and IWSLT 2015 English-Vietnamese machine translation tasks compared with strong NMT baselines. We also observe that our method produces meaningful phrases in output languages.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑