Multilingual LLM-as-a-Judge on Belebele
63.92AccuracySEMJ
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| SEMJ2026.06 | 63.92 | 21.75 | |
| Multilingual Vote2026.06 | 63.7 | — | |
| Monolingual Ref2026.06 | 63.46 | 19.65 | |
| Monolingual Vote2026.06 | 62.54 | 19.14 | |
| Base Judge2026.06 | 62.53 | 19.75 |
| Method | Links | ||
|---|---|---|---|
| SEMJ2026.06 | 63.92 | 21.75 | |
| Multilingual Vote2026.06 | 63.7 | — | |
| Monolingual Ref2026.06 | 63.46 | 19.65 | |
| Monolingual Vote2026.06 | 62.54 | 19.14 | |
| Base Judge2026.06 | 62.53 | 19.75 |