Data-to-text Generation on E2E NLG Challenge Czech (test)
24.87BLEU (sacrebleu)binmt
Evaluation results
| Method | Links | |||||||
|---|---|---|---|---|---|---|---|---|
| binmtpre-training=machine translation2020.04 | 24.87 | 1.9 | 5.24 | 25.81 | 47.07 | 2.6 | 26.35 | |
| nmtpre-training=machine translation2020.04 | 24.41 | 2.38 | 5.19 | 25.46 | 46.85 | 2.55 | 25.84 | |
| tgen-sotaarchitecture=pipelined2020.04 | 20.6 | 2.75 | 4.77 | 23.32 | 42.95 | 2.18 | 21.96 | |
| masspre-training=unsupervised transfer learning2020.04 | 16.61 | 24.82 | 4.22 | 21.16 | 38.94 | 1.75 | 17.72 | |
| scratchtraining=from scratch2020.04 | 11.19 | 63.18 | 3.06 | 15.79 | 28.27 | 0.84 | 11.66 |