Simultaneous Translation Policies: From Fixed to Adaptive

Adaptive policies are better than fixed policies for simultaneous translation, since they can flexibly balance the tradeoff between translation quality and latency based on the current context information. But previous methods on obtaining adaptive policies either rely on complicated training process, or underperform simple fixed policies. We design an algorithm to achieve adaptive policies via a simple heuristic composition of a set of fixed policies. Experiments on Chinese -> English and German -> English show that our adaptive policies can outperform fixed ones by up to 4 BLEU points for the same latency, and more surprisingly, it even surpasses the BLEU score of full-sentence translation in the greedy mode (and very close to beam mode), but with much lower latency.

[1]  Anoop Sarkar,et al.  Prediction Improves Simultaneous Neural Machine Translation , 2018, EMNLP.

[2]  Rico Sennrich,et al.  Neural Machine Translation of Rare Words with Subword Units , 2015, ACL.

[3]  Thomas G. Dietterich.  Multiple Classifier Systems , 2000, Lecture Notes in Computer Science.

[4]  Alexander H. Waibel,et al.  Simultaneous translation of lectures and speeches , 2007, Machine Translation.

[5]  Srinivas Bangalore,et al.  Incremental Segmentation and Decoding Strategies for Simultaneous Translation , 2013, IJCNLP.

[6]  Bill Byrne,et al.  Unfolding and Shrinking Neural Machine Translation Ensembles , 2017, EMNLP.

[7]  Andrej Ljolje,et al.  Segmentation Strategies for Streaming Speech Translation , 2013, HLT-NAACL.

[8]  Salim Roukos,et al.  Bleu: a Method for Automatic Evaluation of Machine Translation , 2002, ACL.

[9]  Srinivas Bangalore,et al.  Real-time Incremental Speech-to-Speech Translation of Dialogs , 2012, NAACL.

[10]  Kyunghyun Cho,et al.  Can neural machine translation do simultaneous translation? , 2016, ArXiv.

[11]  Renjie Zheng,et al.  Simpler and Faster Learning of Adaptive Policies for Simultaneous Translation , 2019, EMNLP.

[12]  Nadir Durrani,et al.  Incremental Decoding and Training Methods for Simultaneous Translation in Neural Machine Translation , 2018, NAACL.

[13]  Renjie Zheng,et al.  Speculative Beam Search for Simultaneous Translation , 2019, EMNLP.

[14]  Lukasz Kaiser,et al.  Attention is All you Need , 2017, NIPS.

[15]  Wei Li,et al.  Monotonic Infinite Lookback Attention for Simultaneous Machine Translation , 2019, ACL.

[16]  Haifeng Wang,et al.  STACL: Simultaneous Translation with Implicit Anticipation and Controllable Latency using Prefix-to-Prefix Framework , 2018, ACL.

[17]  Samy Bengio,et al.  An Online Sequence-to-Sequence Model Using Partial Conditioning , 2015, NIPS.

[18]  Oleksandr Makeyev,et al.  Neural network with ensembles , 2010, The 2010 International Joint Conference on Neural Networks (IJCNN).

[19]  Renjie Zheng,et al.  Simultaneous Translation with Flexible Policy via Restricted Imitation Learning , 2019, ACL.

[20]  Jordan L. Boyd-Graber,et al.  Don't Until the Final Verb Wait: Reinforcement Learning for Simultaneous Machine Translation , 2014, EMNLP.