Machine Translation Meta-evaluation on WMT EN-JA 2025
57.3Acc*EqDMM (CL)
Evaluation results
| Method | Links | ||
|---|---|---|---|
| DMM (CL)Pooling Strategy=Language-Pooled, Model Variant=DMM, Combiner=CL, k=62026.05 | 57.3 | 73.5 | |
| DMM (CL)Pooling Strategy=Language-Separate, Model Variant=DMM, Combiner=CL, k=62026.05 | 57.1 | 71.6 | |
| MM (GP)Pooling Strategy=Language-Pooled, Model Variant=MM, Combiner=GP2026.05 | 57 | 75.7 | |
| DMM (MLP)Pooling Strategy=Language-Separate, Model Variant=DMM, Combiner=MLP, k=62026.05 | 56.8 | 74.3 | |
| MM (MLP)Pooling Strategy=Language-Separate, Model Variant=MM, Combiner=MLP2026.05 | 56.7 | 73.9 | |
| MM (MLP)Pooling Strategy=Language-Pooled, Model Variant=MM, Combiner=MLP2026.05 | 56.6 | 74.1 | |
| DMM (GP)Pooling Strategy=Language-Pooled, Model Variant=DMM, Combiner=GP, k=62026.05 | 56.6 | 72.8 | |
| DMM (MLP)Pooling Strategy=Language-Pooled, Model Variant=DMM, Combiner=MLP, k=62026.05 | 56.2 | 74.9 | |
| Gemini-2.5-Pro2026.05 | 55.9 | 82.5 | |
| MM (GP)Pooling Strategy=Language-Separate, Model Variant=MM, Combiner=GP2026.05 | 55.9 | 76.6 | |
| DMM (GP)Pooling Strategy=Language-Separate, Model Variant=DMM, Combiner=GP, k=62026.05 | 55.1 | 77 | |
| wmt22-comet-da2026.05 | 53.6 | 52.6 | |
| MM (OLS)Pooling Strategy=Language-Separate, Model Variant=MM, Combiner=OLS2026.05 | 53.4 | 64 | |
| MM (OLS)Pooling Strategy=Language-Pooled, Model Variant=MM, Combiner=OLS2026.05 | 49.4 | 46.5 | |
| DMM (OLS)Pooling Strategy=Language-Separate, Model Variant=DMM, Combiner=OLS, k=62026.05 | 49.4 | 45.4 | |
| DMM (OLS)Pooling Strategy=Language-Pooled, Model Variant=DMM, Combiner=OLS, k=62026.05 | 47.9 | 44.1 | |
| gemba2026.05 | 34.1 | 72.5 |