한국어EN
Prehistory — From Neural Networks to the TransformerDzmitry Bahdanau, Kyunghyun Cho, Yoshua BengioKyunghyun ChoYoshua Bengio

Neural Machine Translation by Jointly Learning to Align and Translate

Neural Machine Translation by Jointly Learning to Align and Translate released on arXiv — introduced attention

Why it matters

The paper that solved the fixed-length vector bottleneck with attention. The direct predecessor of the Transformer's self-attention

Source — Primary sources first; secondary reports are labelled “Press”

Paperarxiv.orghttps://arxiv.org/abs/1409.0473

Added in Week 35, 2026 · See that week

Same week

Very Deep Convolutional Networks for Large-Scale Image Recognition (VGG)Paper

Nearby on the same axis

Very Deep Convolutional Networks for Large-Scale Image Recognition (VGG)PaperGoogle publishes Sequence to Sequence Learning with Neural Networks on arXivPaperDeepMind publishes Neural Turing Machines on arXivPaperGoodfellow et al. publish Generative Adversarial Networks on arXivPaper