Interested in Contributing?

Scored Systems

System Submitter System Notes Constraint Run Notes BLEU BLEU-cased TER BEER 2.0 CharactTER
AI-Translation  (Details) pa-translation
PATECH
yes Transformer + back-translation + ensemble + reranking

44.6

44.6

1.074

0.409

0.557

KSAI-system  (Details) tiny
KSAI
yes

42.4

42.4

0.881

0.395

0.617

Baidu-system  (Details) Baidu-MT
Baidu
yes

42.4

42.4

1.056

0.378

0.608

wyx_enzh_mt  (Details) wyx
yes

40.6

40.6

0.992

0.363

0.634

NEU  (Details) NiuTrans
Northeastern University
Ensemble of 8 deep Transformer (30 layers) models + back-translation with beam search + iterative distillation by ensemble teachers + hypothesis combination yes

39.9

39.9

0.960

0.361

0.669

test  (Details) rnn-lstm-gru
no

38.5

38.5

1.970

0.348

0.697

BTRANS  (Details) BTRANS
test2019xxx
yes

38.3

38.3

1.063

0.365

0.632

Wonder Woman  (Details) fansiawang
Personal
ensemble of 4 transformer models with beam 12 yes

36.5

36.5

0.997

0.343

0.743

Wonder Woman  (Details) fansiawang
Personal
ensemble of 4 transformer models with beam 12 yes ensemble and re-rank

36.4

36.4

0.998

0.342

0.743

Wonder Woman  (Details) fansiawang
Personal
ensemble of 4 transformer models with beam 12 yes ensemble and re-rank

36.4

36.3

0.998

0.340

0.741

research baseline model   (Details) research
Machine Translator
baseline model yes

36.2

36.2

0.977

0.346

0.709

Wonder Woman  (Details) fansiawang
Personal
ensemble of 4 transformer models with beam 12 yes ensemble of 4 transformer models with beam 12

36.0

36.0

0.998

0.341

0.746

Wonder Woman  (Details) fansiawang
Personal
ensemble of 4 transformer models with beam 12 yes

36.0

36.0

0.998

0.341

0.747

Wonder Woman  (Details) fansiawang
Personal
ensemble of 4 transformer models with beam 12 yes ensemble of 4 transformer models with beam 12

36.0

36.0

0.998

0.339

0.744

Wonder Woman  (Details) fansiawang
Personal
ensemble of 4 transformer models with beam 12 yes

36.0

35.9

0.998

0.339

0.745

Wonder Woman  (Details) fansiawang
Personal
ensemble of 4 transformer models with beam 12 yes

35.9

35.9

0.997

0.340

0.747

Wonder Woman  (Details) fansiawang
Personal
ensemble of 4 transformer models with beam 12 yes

35.9

35.9

0.998

0.339

0.744

Wonder Woman  (Details) fansiawang
Personal
ensemble of 4 transformer models with beam 12 yes

35.8

35.8

0.997

0.341

0.753

xzl-sub-nmt  (Details) qianqian
personal
sub yes

35.7

35.7

0.999

0.342

failed

Wonder Woman  (Details) fansiawang
Personal
ensemble of 4 transformer models with beam 12 yes

35.7

35.7

0.997

0.340

0.752

xzl-nmt  (Details) qianqian
personal
fast-align, corpus filtering,transformer-based ensemble, re-ranking, post-processing yes main system

35.6

35.6

0.997

0.342

0.744

Wonder Woman  (Details) fansiawang
Personal
ensemble of 4 transformer models with beam 12 yes

35.6

35.6

0.997

0.341

0.758

Captain Marvel  (Details) marvel
S.H.I.E.L.D
yes

35.5

35.5

failed

0.343

0.703

Wonder Woman  (Details) fansiawang
Personal
ensemble of 4 transformer models with beam 12 yes

35.5

35.5

0.998

0.331

0.729

base  (Details) Translator
yes

35.5

35.5

0.946

0.354

0.714

base  (Details) Translator
yes

35.4

35.4

0.999

0.337

0.724

Wonder Woman  (Details) fansiawang
Personal
ensemble of 4 transformer models with beam 12 yes ensemble of 4 transformer models with beam 12

35.3

35.3

0.997

0.340

0.760

test_online  (Details) abelcui
no

35.3

35.3

11.763

0.190

failed

base  (Details) Translator
yes

35.2

35.2

0.999

0.336

0.726

Wonder Woman  (Details) fansiawang
Personal
ensemble of 4 transformer models with beam 12 yes ensemble of 4 transformer models with beam 12

35.1

35.1

0.997

0.340

0.763

base  (Details) Translator
yes

34.9

34.9

0.999

0.335

0.726

test123  (Details) serenade
Peking University
for test yes

34.8

34.8

failed

0.335

0.756

test123  (Details) serenade
Peking University
for test yes

34.8

34.8

failed

0.335

0.756

test123  (Details) serenade
Peking University
for test yes

34.7

34.7

failed

0.334

0.755

test123  (Details) serenade
Peking University
for test yes

34.6

34.6

failed

0.334

0.762

test123  (Details) serenade
Peking University
for test yes

34.6

34.6

failed

0.334

0.762

toy-nmt  (Details) serenade
Peking University
Single Model yes

34.6

34.6

failed

0.334

0.762

test_big_data  (Details) abelcui
no with bt but no parallel and convert punct

34.6

34.5

12.694

0.191

0.796

NICT  (Details) Nedved
NICT
Transformer, back-translation, ensemble, and fine-tuning yes

34.5

34.5

0.999

0.326

0.734

base  (Details) Translator
yes

34.5

34.4

1.000

0.334

0.738

UEDIN  (Details) XapaJIaMnu
UEDIN
yes Baseline system

34.4

34.4

1.010

0.337

0.749

base  (Details) Translator
yes

34.2

34.2

0.999

0.332

0.759

base  (Details) Translator
yes

34.2

34.2

0.999

0.332

0.759

base  (Details) Translator
yes

34.2

34.2

1.000

0.332

0.763

test123  (Details) serenade
Peking University
for test yes

34.1

34.1

failed

0.336

0.764

ForyorMT_English2Chinese  (Details) Zeng Hui
Single base transformer + back-translation of 10 million general news corpus no Single base transformer + back-translation of 10 million general news corpus

34.1

34.1

12.049

0.196

0.794

test123  (Details) serenade
Peking University
for test yes

34.1

34.1

failed

0.330

0.774

ForyorMT_English2Chinese  (Details) Zeng Hui
Single base transformer model + 1 million back-translated sentence pairs + checkpoint averaging yes Single base transformer model + 1 million back-translated sentence pairs + checkpoint averaging

34.0

34.0

12.044

0.198

0.789

base  (Details) Translator
yes

34.0

34.0

11.730

0.196

0.795

base  (Details) Translator
yes

34.0

34.0

0.994

0.341

0.759

ForyorMT_English2Chinese  (Details) Zeng Hui
Single base transformer model + 1 million back-translated sentence pairs yes Single base transformer model + 1 million back-translated sentence pairs

33.8

33.8

12.015

0.197

0.795

base  (Details) Translator
yes

33.3

33.2

1.000

0.330

0.776

test_big_data  (Details) abelcui
no

33.2

33.2

1.000

0.330

0.807

test_big_data  (Details) abelcui
no

33.2

33.2

11.267

0.200

0.810

PROMT NMT EN-ZH  (Details) Alex Molchanov
PROMT LLC
no

33.0

33.0

1.011

0.345

0.739

test_big_data  (Details) abelcui
no add parallel and bt

32.4

32.4

12.644

0.190

0.803

test123  (Details) serenade
Peking University
for test yes

32.2

32.2

failed

0.322

0.801

IIE.STRANS.EN2ZH  (Details) Xiangpeng Wei
IIE, CAS
yes

32.1

32.1

1.110

0.338

0.738

ForyorMT  (Details) Zeng Hui
Single base transformer model. yes Single base transformer model.

32.1

32.1

11.621

0.195

0.811

test123  (Details) serenade
Peking University
for test yes

30.0

30.0

failed

0.312

0.881

test123  (Details) serenade
Peking University
for test yes

29.9

29.9

failed

0.312

0.882