Peer-reviewed research
BLEU: a Method for Automatic Evaluation of Machine Translation
Kishore Papineni et al., Proceedings of the 40th Annual Meeting of the Association for Computational Linguistics, 2002
Read the originalRead on October 17, 2026
- What it is
- The paper that introduced BLEU, the automatic score most speech translation papers use. It proposes automatic evaluation that is quick, inexpensive and language-independent, and that correlates with human judgement.
- How much weight it carries
- A peer-reviewed, widely cited paper from IBM Research. BLEU measures overlap with reference translations, so it can miss meaning that is correct but worded differently.
Facts taken from this source
- The 2002 paper that introduced BLEU proposed a quick, inexpensive, language-independent automatic method that correlates highly with human evaluation.
“that is quick, inexpensive, and language-independent, that correlates highly with human” (Abstract)