Sequence-to-sequence

Neural sequence transformation was applied to translation.

Why it matters

It demonstrated end-to-end translation with paired recurrent networks.

The reported scores are for the stated WMT task, not a general measure of language understanding.

Event date

September 10, 2014

Timeline date

Event date

Verification

Sources gathered automatically · September 11, 2026

Lines

ID

evt-0008

primary Sequence to Sequence Learning
arXiv · Published September 10, 2014
primary Sequence to Sequence Learning with Neural Networks
Advances in Neural Information Processing Systems 27 (NIPS 2014), NeurIPS proceedings · Published 2014

Related Attention for translation 2014
Two papers from the same month on neural sequence translation.
Builds on LSTM 1997
The authors explicitly use multilayer LSTM networks to encode the input sequence and generate a translation.
- Sequence to Sequence Learning · arXiv

Builds on Show and Tell: a caption for an image 2014links to this record
The model is carried over from encoder-decoder machine translation (Sutskever, Vinyals, Le): an image is encoded in place of a source sentence.
- Show and Tell: A Neural Image Caption Generator (arXiv:1411.4555v1) · arXiv (Google)
Builds on GNMT: neural translation in Google Translate 2016links to this record
The post explains GNMT as recurrent networks learning a mapping from an input sequence to an output sequence directly, citing Sutskever, Vinyals and Le 2014.
- A Neural Network for Machine Translation, at Production Scale · Google Research