Neural Machine Translation by Jointly Learning to Align and Translate
arXiv: 1409.0473
TLDR (English)
The seminal attention mechanism paper (pre-Transformer). The authors found that seq2seq's fixed-length bottleneck vector limited translation quality, and proposed letting the decoder dynamically attend to all encoder hidden states when generating each word. This idea directly evolved into Transformer self-attention.
TLDR（中文）
注意力机制的开山之作（在 Transformer 之前）。作者发现 Seq2Seq 的固定长度瓶颈向量限制了翻译 质量，提出让解码器在生成每个词时都能"回顾"编码器的所有隐状态，动态分配注意力权重。 这个思想直接演化为 Transformer 的自注意力。