Interested in Contributing?
- Check out available resources.
- Create an account and start submitting your own systems.
Scored Systems
| System | Submitter | System Notes | Run Notes | BLEU | BLEU (11b) | BLEU-cased | BLEU-cased (11b) | TER |
|---|---|---|---|---|---|---|---|---|
| DCU_En--Es_Primary system (Details) | DCU CNGL, Dublin City University |
System Combination of PB-SMT, Factored model, Apertitum RBMT and Marclator (Marker based EBMT system). | CNGL, School of Computing, Dublin City University, English to Spanish PRIMARY SUBMISSION (constrained track) |
30.4 |
28.4 |
0.587 |
||
| prueba (Details) | pakozm Universidad CEU-Cardenal Herrera and Universitat Politécnica de València |
30.2 |
28.3 |
0.588 |
||||
| DCU_En--Es_constrained-1 (Details) | DCU CNGL, Dublin City University |
Training data: Europarl, news commentary and part of UN data | English to Spanish constrained track |
29.8 |
27.7 |
0.596 |
||
| uedin-wmt10-en-es (Details) | Edinburgh University of Edinburgh |
constraint, no LDC data |
29.6 |
28.6 |
0.591 |
|||
| DCU_En--Es_constrained-2 (Details) | DCU CNGL, Dublin City University |
Training data: All of Europarl, news commentary and UN data | English to Spanish constrained track |
29.6 |
27.6 |
0.597 |
||
| DCU_En--Es_constrained-3 (Details) | DCU CNGL, Dublin City University |
Training data: Europarl and news commentary. | English to Spanish constrained track. |
29.4 |
27.4 |
0.596 |
||
| DCU_En--Es_constrained-4 (Details) | DCU CNGL, Dublin City University |
Training data: All. Using MOSES and EBMT phrases. | English to Spanish constrained track. |
29.4 |
27.3 |
0.601 |
||
| prueba (Details) | pakozm Universidad CEU-Cardenal Herrera and Universitat Politécnica de València |
29.2 |
27.4 |
0.600 |
||||
| Cambridge WMT10 EN-ES (Details) | Gonzalo Iglesias Cambridge |
29.1 |
27.8 |
0.607 |
||||
| UPV-PRHLT (Details) | gsanchis Departamento de Sistemas Informáticos y Computación - Instituto Tecnológico de Informática - Universidad Politécnica de Valencia |
28.7 |
26.9 |
0.604 |
||||
| UCH-UPV NGram-Based SMT (Details) | Francisco Zamora Martínez Universidad CEU-Cardenal Herrera and Universidad Politécnica de Valencia |
NGram-based SMT system developed using our own decoder. Trained over all available corpora for Spanish, and all available bilingual corpora for english-spanish. |
28.5 |
26.6 |
0.608 |
|||
| JHU en-es (Details) | dowobeha Johns Hopkins University |
Primary submission. No LDC data. |
28.1 |
26.5 |
0.596 |
|||
| UCH-UPV (Details) | Francisco Zamora Martínez Universidad CEU-Cardenal Herrera and Universidad Politécnica de Valencia |
Moses n-best list output reranked with a Neural Network Language Model (NNLM). The NNLM is trained with all the news corpora. |
28.0 |
26.1 |
0.610 |
|||
| SFU (Details) | baskarans Simon Fraser University |
In-house implementation of Hiero | Constrained; ed LM |
24.3 |
23.2 |
0.620 |
||
| DFKI's hybrid translation system for WMT10 (Details) | cfedermann DFKI GmbH |
23.7 |
22.8 |
0.650 |
||||
| SFU (Details) | baskarans Simon Fraser University |
In-house implementation of Hiero | Constrained; UN data not used in LM generation |
22.8 |
21.9 |
0.621 |
||
| KU_TRBMT_L1 (Details) | ebicici Koc University |
L1 Regularized Transductive Regression Based Machine Translation reranking of Moses output |
21.4 |
18.9 |
0.675 |
|||
| uk-dan (Details) | zeman Charles University in Prague, ÚFAL |
Trained on News Commentary only. |
21.1 |
16.5 |
0.687 |