Machine Translation on WMT22 Liv-En (test)
43.6COMET ScoreDUAL-REFLECT
Evaluation results
| Method | Links | ||
|---|---|---|---|
| DUAL-REFLECTInference Methodology=DUAL-REFLECT2024.06 | 43.6 | 41.7 | |
| ChatGPT + MADInference Methodology=MAD2024.06 | 43.2 | 41.3 | |
| ChatGPT + 5-shotPrompting Strategy=Few-shot, Shots=52024.06 | 42.7 | 40.9 | |
| ChatGPTPrompting Strategy=Zero-shot2024.06 | 40.6 | 41.1 |