Wu and colleagues: Google neural machine translation system

What changed

Described and evaluated an attention-based deep neural translation system with wordpieces and inference optimizations deployed at production scale.

Scientific scope

Described and evaluated an attention-based deep neural translation system with wordpieces and inference optimizations deployed at production scale. Scope limitation: The GNMT system combines existing and new engineering choices; it is not the invention of neural machine translation or attention.

Evidence

The retained primary-source abstract, bibliographic record and stated review scope support this provisional assessment. No independent experimental replication by WRS is claimed. The GNMT system combines existing and new engineering choices; it is not the invention of neural machine translation or attention.

Contributors

• The named authors listed in the linked research record · SCIENCE: Complete source byline retained in the research record; human research direction, method development and evaluation. Individual labor shares are not measured.

• Wu and colleagues: Google neural machine translation system — original paper
• Wu and colleagues: Google neural machine translation system — official company source

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