Machine Translation Meta-evaluation on WMT EN-UK 2025
0.572Acc*Eqwmt22-comet-da
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| wmt22-comet-da2026.05 | 0.572 | 0.696 | |
| MM (MLP)Pooling Strategy=Language-Separate, Model Variant=MM, Combiner=MLP2026.05 | 0.557 | 0.715 | |
| MM (MLP)Pooling Strategy=Language-Pooled, Model Variant=MM, Combiner=MLP2026.05 | 0.556 | 0.692 | |
| DMM (GP)Pooling Strategy=Language-Pooled, Model Variant=DMM, Combiner=GP, k=62026.05 | 0.555 | 0.664 | |
| DMM (MLP)Pooling Strategy=Language-Separate, Model Variant=DMM, Combiner=MLP, k=62026.05 | 0.555 | 0.682 | |
| DMM (MLP)Pooling Strategy=Language-Pooled, Model Variant=DMM, Combiner=MLP, k=62026.05 | 0.554 | 0.679 | |
| DMM (GP)Pooling Strategy=Language-Separate, Model Variant=DMM, Combiner=GP, k=62026.05 | 0.551 | 0.748 | |
| MM (GP)Pooling Strategy=Language-Separate, Model Variant=MM, Combiner=GP2026.05 | 0.548 | 0.752 | |
| DMM (CL)Pooling Strategy=Language-Pooled, Model Variant=DMM, Combiner=CL, k=62026.05 | 0.547 | 0.639 | |
| MM (GP)Pooling Strategy=Language-Pooled, Model Variant=MM, Combiner=GP2026.05 | 0.543 | 0.718 | |
| DMM (CL)Pooling Strategy=Language-Separate, Model Variant=DMM, Combiner=CL, k=62026.05 | 0.543 | 0.634 | |
| Gemini-2.5-Pro2026.05 | 0.517 | 0.819 | |
| MM (OLS)Pooling Strategy=Language-Separate, Model Variant=MM, Combiner=OLS2026.05 | 0.488 | 0.44 | |
| MM (OLS)Pooling Strategy=Language-Pooled, Model Variant=MM, Combiner=OLS2026.05 | 0.472 | 0.372 | |
| DMM (OLS)Pooling Strategy=Language-Separate, Model Variant=DMM, Combiner=OLS, k=62026.05 | 0.467 | 0.315 | |
| DMM (OLS)Pooling Strategy=Language-Pooled, Model Variant=DMM, Combiner=OLS, k=62026.05 | 0.464 | 0.364 | |
| gemba2026.05 | 0.346 | 0.741 |