Multilingual LLM-as-a-Judge on Wino-X
72.92AccuracySEMJ
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| SEMJ2026.06 | 72.92 | 39.22 | |
| Monolingual Ref2026.06 | 68.75 | 29.35 | |
| Monolingual Vote2026.06 | 67.86 | 22.68 | |
| Base Judge2026.06 | 66.96 | 21.52 | |
| Multilingual Vote2026.06 | 50.44 | — |
| Method | Links | ||
|---|---|---|---|
| SEMJ2026.06 | 72.92 | 39.22 | |
| Monolingual Ref2026.06 | 68.75 | 29.35 | |
| Monolingual Vote2026.06 | 67.86 | 22.68 | |
| Base Judge2026.06 | 66.96 | 21.52 | |
| Multilingual Vote2026.06 | 50.44 | — |