Papineni, Roukos, Ward and Zhu: BLEU machine-translation evaluation

What changed

Defined corpus BLEU from clipped n-gram precision and brevity penalty with reported human-correlation checks.

Scientific scope

The record is the exact four-author ACL 2002 paper. BLEU measures reference overlap rather than meaning, depends on tokenization, references and aggregation choices, and is not a universal substitute for human or task-specific evaluation.

Evidence

The linked sources identify this contribution and its scope. WRS has not independently reconstructed the scientific claims. The record is the exact four-author ACL 2002 paper. BLEU measures reference overlap rather than meaning, depends on tokenization, references and aggregation choices, and is not a universal substitute for human or task-specific evaluation.

Contributors

• Kishore Papineni · SCIENCE: Named source author of the scoped work; individual effort shares are not measured.
• Salim Roukos · SCIENCE: Named source author of the scoped work; individual effort shares are not measured.
• Todd Ward · SCIENCE: Named source author of the scoped work; individual effort shares are not measured.
• Wei-Jing Zhu · SCIENCE: Named source author of the scoped work; individual effort shares are not measured.

• ACL Anthology record and paper for BLEU
• DBLP bibliography for BLEU

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