# |
Team |
Task |
Date/Time |
DataID |
AMFM |
Method
|
Other Resources
|
System Description |
unuse |
unuse |
unuse |
unuse |
unuse |
unuse |
unuse |
unuse |
unuse |
unuse |
|
1 | NLPHut | INDIC21en-ta | 2021/03/20 00:16:59 | 4616 | - | - | - | - | - | - | 0.739011 | - | - | - | NMT | No | Transformer with target language tag trained using all languages PMI data. Then fine tuned using en-ta PMI data.
|
2 | ORGANIZER | INDIC21en-ta | 2021/04/08 17:25:38 | 4804 | - | - | - | - | - | - | 0.723160 | - | - | - | NMT | No | Bilingual baseline trained on PMI data. Transformer base. LR=10-3 |
3 | NICT-5 | INDIC21en-ta | 2021/04/21 15:45:41 | 5289 | - | - | - | - | - | - | 0.776138 | - | - | - | NMT | No | Pretrain MBART on IndicCorp and FT on bilingual PMI data. Beam search. Model is bilingual. |
4 | SRPOL | INDIC21en-ta | 2021/04/21 19:22:28 | 5323 | - | - | - | - | - | - | 0.795654 | - | - | - | NMT | No | Base transformer on all WAT21 data |
5 | NICT-5 | INDIC21en-ta | 2021/04/22 11:54:02 | 5364 | - | - | - | - | - | - | 0.792622 | - | - | - | NMT | No | MBART+MNMT. Beam 4. |
6 | gaurvar | INDIC21en-ta | 2021/04/25 20:02:43 | 5586 | - | - | - | - | - | - | 0.694376 | - | - | - | NMT | No | Multi Task Multi Lingual T5 trained for Multiple Indic Languages |
7 | sakura | INDIC21en-ta | 2021/05/01 11:37:07 | 5890 | - | - | - | - | - | - | 0.791070 | - | - | - | NMT | No | Fine-tuning of multilingual mBART one2many model with training corpus.
|
8 | gaurvar | INDIC21en-ta | 2021/05/01 19:34:16 | 5934 | - | - | - | - | - | - | 0.684232 | - | - | - | NMT | No | Multi Task Multi Lingual T5 trained for Multiple Indic Languages |
9 | mcairt | INDIC21en-ta | 2021/05/03 17:06:58 | 5995 | - | - | - | - | - | - | 0.801632 | - | - | - | NMT | No | multilingual model(one to many model) trained on all WAT 2021 data by using base transformer. |
10 | IIIT-H | INDIC21en-ta | 2021/05/03 18:11:11 | 6013 | - | - | - | - | - | - | 0.778991 | - | - | - | NMT | No | MNMT system (En-XX) trained via exploiting lexical similarity on PMI+CVIT parallel corpus, then improved using back translation on PMI monolingual data followed by fine tuning. |
11 | CFILT | INDIC21en-ta | 2021/05/04 01:06:43 | 6050 | - | - | - | - | - | - | 0.802920 | - | - | - | NMT | No | Multilingual(One-to-Many(En-XX)) NMT model based on Transformer with shared encoder and decoder. |
12 | coastal | INDIC21en-ta | 2021/05/04 01:40:19 | 6086 | - | - | - | - | - | - | 0.788022 | - | - | - | NMT | No | seq2seq model trained on all WAT2021 data |
13 | sakura | INDIC21en-ta | 2021/05/04 04:18:16 | 6159 | - | - | - | - | - | - | 0.795712 | - | - | - | NMT | No | Pre-training multilingual mBART one2many model with training corpus followed by finetuning on PMI Parallel.
|
14 | SRPOL | INDIC21en-ta | 2021/05/04 15:22:51 | 6240 | - | - | - | - | - | - | 0.798837 | - | - | - | NMT | No | Ensemble of one-to-many on all data. Pretrained on BT, finetuned on PMI |
15 | SRPOL | INDIC21en-ta | 2021/05/04 16:28:29 | 6266 | - | - | - | - | - | - | 0.799382 | - | - | - | NMT | No | One-to-many on all data. Pretrained on BT, finetuned on PMI |
16 | IITP-MT | INDIC21en-ta | 2021/05/04 18:14:43 | 6303 | - | - | - | - | - | - | 0.756693 | - | - | - | NMT | No | One-to-Many model trained on all training data with base Transformer. All indic language data is romanized. Model fine-tuned on BT PMI monolingual corpus. |
17 | NICT-5 | INDIC21en-ta | 2021/06/25 11:39:29 | 6491 | - | - | - | - | - | - | 0.000000 | - | - | - | NMT | No | Using PMI and PIB data for fine-tuning on a mbart model trained for over 5 epochs. MNMT model. |