Bahdanau, Cho and Bengio: jointly learning to align and translate

What changed

Neural translation could dynamically select source context instead of relying on one fixed vector.

Scientific scope

2014 preprint/2015 ICLR paper in neural translation; not every later attention mechanism or Transformer architecture.

Evidence

The linked sources identify this contribution and its scope. WRS has not independently reconstructed the scientific claims. The record is the 2014 preprint later presented at ICLR 2015. It concerns learned alignment in neural machine translation, not every later mechanism called attention.

Contributors

• Dzmitry Bahdanau · SCIENCE: Named source author of the scoped work; individual effort shares are not measured.
• Kyunghyun Cho · SCIENCE: Named source author of the scoped work; individual effort shares are not measured.
• Yoshua Bengio · SCIENCE: Named source author of the scoped work; individual effort shares are not measured.

• Bahdanau, Cho and Bengio, Neural Machine Translation by Jointly Learning to Align and Translate

Provisional WRS assessment, edition 2026-10-16.1. Independent expert calibration is pending. This grade does not certify correctness or assign ownership.

Full scientific record · Browse contribution grades · Request a correction