Machine Translation on FLORES-200 uig_Arab-English (test)
86.4COMET-22SCALE-refine
Evaluation Results
| Method | Links | ||||
|---|---|---|---|---|---|
| SCALE-refineLLM=GPT-4, STM=NLLB-3.3B2023.09 | 86.4 | 85 | 72.2 | 27.9 | |
| NLLBModel Version=3.3B, Evaluation Protocol=Supervised2023.09 | 85.4 | 84.4 | 70.4 | 27.5 | |
| GPT-4Evaluation Protocol=few-shot2023.09 | 83.7 | 82.8 | 67.4 | 23.1 | |
| MicrosoftType=Commercial System2023.09 | 82.7 | 81.7 | 66.2 | 21.6 | |
| GPT-3.5Evaluation Protocol=few-shot2023.09 | 73.7 | 74.2 | 53 | 11.6 | |
| XGLMEvaluation Protocol=few-shot2023.09 | 37.1 | 52.8 | 16.9 | 0.2 |