Categories Companies Research

Loading the SOTA2 catalog…

Explore

Categories Companies Research

Contribute

Join SOTA2 Contact

SOTA2

© 2026 SOTA2

Build in Bellevue, USA

WizWand is now SOTA2 Research

Papers Benchmarks Datasets Tasks

Machine Translation research benchmarks – Page 35 · SOTA2 Research

Machine Translation

Benchmarks

Dataset NameSOTA MethodSort

Dataset NameSOTA methodMetricTrendResultsLast Updated

newstest En-Ru 2013 (dev)

Base BPE-level NMT (Row k)

21.17BLEU

6

Feb 26, 2026

newstest En-Cs 2013 (dev)

Bi-scale Character-level NMT (Row j)

0.198BLEU

6

Feb 26, 2026

IWSLT16 En-De (dev)

29.7BLEU

6

Feb 26, 2026

WMT 2014 (newstest14)

26.1BLEU

6

Feb 26, 2026

newstest English-Russian 2015 (test)

24.3BLEU (single)

6

Feb 26, 2026

WMT English-to-French No UNK 2014 (test)

36.15BLEU

6

Feb 26, 2026

WMT En-Zh 2020 (test)

44.3SacreBLEU

6

Feb 26, 2026

WMT English-Chinese 2024

Claude-3.5-Sonnet

99.3ConsistencyChecker

6

Feb 26, 2026

WMT Czech-Ukrainian 2024

Claude-3.5-Sonnet

98.1ConsistencyChecker

6

Feb 26, 2026

IWSLT Speech 2017

84.7sCMT Score

5

May 29, 2026

Guofeng Literary V1

76.2sCMT Score

5

May 29, 2026

Discovered Challenge Topics

Victim Impact Statements

21Difficulty Score

5

May 8, 2026

21.6BLEU

5

Apr 14, 2026

20BLEU

5

Apr 14, 2026

25.7BLEU

5

Apr 14, 2026

25.8BLEU

5

Apr 14, 2026

31.4BLEU

5

Apr 14, 2026

49.4BLEU

5

Apr 14, 2026

Americas Indigenous Languages

40F-score

5

Mar 27, 2026

Vietnamese-Bahnar translation (test)

Sentence boundary

41.33BLEU Score

5

Feb 26, 2026

Bangla conversational text

57BLEU

5

Feb 26, 2026

33.34BLEU

5

Feb 26, 2026

Multilingual Translation

31.9Score

5

Feb 26, 2026

FLoRes-200 Japanese (test)

Vanilla Decoding

26.3BLEU

5

Feb 26, 2026

FLoRes-200 Korean (test)

21.7BLEU

5

Feb 26, 2026

Page 35 of 58

Popular datasets

WMT FLORES IWSLT newstest Flores-200

Follow for updates

Join SOTA2 Contact