Neural Phrase-based Machine Translation

By Po-Sen Huang, Chong Wang, Dengyong Zhou, Li Deng. 2017

25 citations recorded by the source corpus.

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

1. Xu Tan
2. Tao Qin
3. Myle Ott
4. Lijun Wu
5. Tie-Yan Liu
6. Yingce Xia
7. Jiatao Gu
8. Zhe Gan

Citations in the corpus

• Classical Structured Prediction Losses for Sequence to Sequence Learning (2017, background): “We also show the likelihood baseline (MLE) of Wiseman and Rush (2016), their beam search optimization method (BSO), the actor critic result of Bahdanau et al. (2016) as well as the best reported result on this dataset to date by Huang et al. (2017).”
• Efficient Bidirectional Neural Machine Translation (2019, methodology): “96 NPMT+LM (Huang et al., 2017) 29.”
• Layer-Wise Coordination between Encoder and Decoder for Neural Machine Translation (2018, methodology): “21.83 AC+LL [1] 28.53 NPMT [11] 28.96 Dual Transfer Learning [34] 32”
• Learning to Remember Translation History with a Continuous Cache (2017, background): “The latter is especially encouraging to us, since phrases play an important role in machine translation while it is difficult to integrate them into current NMT models (Zhou et al., 2017; Wang et al., 2017c; Huang et al., 2017).”
• Double Path Networks for Sequence to Sequence Learning (2018, methodology): “, 2016) and a phrased-based model with an auxiliary language model (Huang et al., 2017) separately.”
• Multi-layer Representation Fusion for Neural Machine Translation (2018, methodology): “ we restrict ourselves to Transformer for our experiments, Eq. (3) actually shows a more general case. In a sense, our model is doing something similar to fully connected networks in computer vision (Huang et al., 2017a; Liu et al., 2018). Representation fusion is essentially a process that creates ”
• Sentence-wise Smooth Regularization for Sequence to Sequence Learning (2018, methodology): “We use the concatenation of dev2010, tst2010, tst2011 and tst2011 as the test set, which is widely adopted in (Huang et al. 2017; Bahdanau et al. 2016).”
• Online Distilling from Checkpoints for Neural Machine Translation (2019, background): “Following the settings in Huang et al. (2017), the Vietnamese and English vocabulary size are 7,709 and 17,191, respectively.”