Unsupervised Machine Translation On Wmt2016 3
Metriken
BLEU
Ergebnisse
Leistungsergebnisse verschiedener Modelle zu diesem Benchmark
| Paper Title | ||
|---|---|---|
| GPT-3 175B (Few-Shot) | 39.5 | Language Models are Few-Shot Learners |
| MASS (6-layer Transformer) | 33.1 | MASS: Masked Sequence to Sequence Pre-training for Language Generation |
| MLM pretraining for encoder and decoder | 31.8 | Cross-lingual Language Model Pretraining |
0 of 3 row(s) selected.