大體思想和RNN encoder-decoder是一樣的，只是用來LSTM來實現。

paper提到三個important point：

1）encoder和decoder的LSTM是兩個不同的模型

2）deep LSTM表現比shallow好，選用了4層的LSTM

3）實踐中發現將輸入句子reverse後再進行訓練效果更好。So for example, instead of mapping the sentence a,b,c to the sentence α,β,γ, the LSTM is asked to map c,b,a to α,β,γ, where α, β, γ is the translation of a, b, c. This way, a is in close proximity to α, b is fairly close to β, and so on, a fact that makes it easy for SGD to “establish communication” between the input and the output.

論文筆記-Sequence to Sequence Learning with Neural Networks