Multilingual LLM-as-a-Judge on XCOPA
63.36AccuracySEMJ
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| SEMJ2026.06 | 63.36 | 0.447 | |
| Monolingual Ref2026.06 | 60.69 | 0.4211 | |
| Monolingual Vote2026.06 | 59 | 0.4349 | |
| Base Judge2026.06 | 58.64 | 0.4049 | |
| Multilingual Vote2026.06 | 49 | — |
| Method | Links | ||
|---|---|---|---|
| SEMJ2026.06 | 63.36 | 0.447 | |
| Monolingual Ref2026.06 | 60.69 | 0.4211 | |
| Monolingual Vote2026.06 | 59 | 0.4349 | |
| Base Judge2026.06 | 58.64 | 0.4049 | |
| Multilingual Vote2026.06 | 49 | — |