Machine Translation on WMT22 En-Liv (test)
56COMET ScoreDUAL-REFLECT
Evaluation results
| Method | Links | ||
|---|---|---|---|
| DUAL-REFLECTInference Methodology=DUAL-REFLECT2024.06 | 56 | 43.3 | |
| ChatGPT + MADInference Methodology=MAD2024.06 | 55.5 | 42.5 | |
| ChatGPT + 5-shotPrompting Strategy=Few-shot, Shots=52024.06 | 55.3 | 42.1 | |
| ChatGPTPrompting Strategy=Zero-shot2024.06 | 52.7 | 41.8 |