attention总结

1.Show, Attend and Tell: Neural Image Caption Generation with Visual Attention

提出了两种 attention 模式，即 hard attention 和 soft attention

2.Effective Approaches to Attention-based Neural Machine Translation

文章提出了两种 attention 的改进版本，即 global attention 和 local attention。

3.Attention Is All You Need

提出self attention

4.Hierarchical Attention Networks for Document Classification

提出了Hierarchical Attention用于文档分类

5.Attention-over-Attention Neural Networks for Reading Comprehension

提出了Attention Over Attention的Attention机制

6.Convolutional Sequence to Sequence Learning

论文中还采用了 Multi-step Attention