Machine Translation on WMT 22 (test)
85.05COMETWMT22 Winners
Evaluation results
| Method | Links | ||||||
|---|---|---|---|---|---|---|---|
| WMT22 Winners2024.06 | 85.05 | 39.98 | — | — | — | — | |
| TASTEBackbone=LLaMA-2-7b, Training Strategy=Tuning with Fixed Embedding Layer, Subtask Form=Text Classification2024.06 | 82.92 | 29.7 | — | — | — | — | |
| TASTEBackbone=LLaMA-2-7b, Training Strategy=Tuning with Fixed Embedding Layer, Subtask Form=Quality Estimation2024.06 | 82.86 | 29.37 | — | — | — | — | |
| MT-FixEmbBackbone=LLaMA-2-7b, Training Strategy=Tuning with Fixed Embedding Layer2024.06 | 82.59 | 29 | — | — | — | — | |
| TASTEBackbone=LLaMA-2-7b, Training Strategy=Full-Parameter Tuning, Subtask Form=Quality Estimation2024.06 | 82.57 | 29.04 | — | — | — | — | |
| TASTEBackbone=LLaMA-2-7b, Training Strategy=Full-Parameter Tuning, Subtask Form=Text Classification2024.06 | 82.55 | 28.91 | — | — | — | — | |
| MT-FullBackbone=LLaMA-2-7b, Training Strategy=Full-Parameter Tuning2024.06 | 82.39 | 28.52 | — | — | — | — | |
| NLLB-3.3b2024.06 | 82.03 | 29.28 | — | — | — | — | |
| BaylingBackbone=LLaMA-2-7b2024.06 | 81.82 | 28.08 | — | — | — | — | |
| TASTEBackbone=BLOOMZ-7b1-mt, Training Strategy=Tuning with Fixed Embedding Layer, Subtask Form=Quality Estimation2024.06 | 80.43 | 27.96 | — | — | — | — | |
| TASTEBackbone=BLOOMZ-7b1-mt, Training Strategy=Tuning with Fixed Embedding Layer, Subtask Form=Text Classification2024.06 | 80.38 | 27.88 | — | — | — | — | |
| ParroTBackbone=LLaMA-2-7b2024.06 | 80.05 | 25.98 | — | — | — | — | |
| TIMBackbone=BLOOMZ-7b1-mt2024.06 | 79.67 | 27.34 | — | — | — | — | |
| TASTEBackbone=BLOOMZ-7b1-mt, Training Strategy=Full-Parameter Tuning, Subtask Form=Text Classification2024.06 | 79.59 | 26.47 | — | — | — | — | |
| TASTEBackbone=BLOOMZ-7b1-mt, Training Strategy=Full-Parameter Tuning, Subtask Form=Quality Estimation2024.06 | 79.56 | 26.51 | — | — | — | — | |
| MT-FixEmbBackbone=BLOOMZ-7b1-mt, Training Strategy=Tuning with Fixed Embedding Layer2024.06 | 78.84 | 26.15 | — | — | — | — | |
| ParroTBackbone=BLOOMZ-7b1-mt2024.06 | 78.53 | 25.65 | — | — | — | — | |
| MT-FullBackbone=BLOOMZ-7b1-mt, Training Strategy=Full-Parameter Tuning2024.06 | 78.3 | 25.3 | — | — | — | — | |
| ChatGPT + Dual ReflectionBase Model=ChatGPT, Strategy=Dual Reflection2024.06 | — | — | 82.4 | 84.7 | 84.2 | 83.8 | |
| ChatGPT + RerankBase Model=ChatGPT, Strategy=Rerank2024.06 | — | — | 82.1 | 84.4 | 83.6 | 83 | |
| ChatGPT + Self-ReflectBase Model=ChatGPT, Strategy=Self-Reflect2024.06 | — | — | 82 | 84.4 | 83.3 | 83.1 |