Machine Translation on WMT23 En-Zh (test)
47.5BLEUX-ALMA-13B (Only SFT, MoE)
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| X-ALMA-13B (Only SFT, MoE)Parameters=13B, Training Strategy=Only SFT, Architecture=MoE2025.02 | 47.5 | 86.1 | |
| Aya-23-8BParameters=8B2025.02 | 44.5 | 85.3 | |
| Aya-23-35BParameters=35B2025.02 | 42.8 | 84.6 | |
| X-ALMA-13B-DAT (MoE)Parameters=13B, Training Strategy=DAT, Architecture=MoE2025.02 | 40.3 | 85 | |
| X-ALMA-13B-DATM (MoE)Parameters=13B, Training Strategy=DATM, Architecture=MoE2025.02 | 40.3 | 85 | |
| NLLB-3.3BParameters=3.3B2025.02 | 34.8 | 79.6 | |
| LLaMAX3-Alpaca-8BParameters=8B2025.02 | 34 | 81.5 | |
| Aya-1012025.02 | 25.4 | 78.6 |