Multilingual LLM-as-a-Judge on XStoryCloze
78.84AccuracySEMJ
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| SEMJ2026.06 | 78.84 | 32.07 | |
| Monolingual Ref2026.06 | 75.56 | 21 | |
| Multilingual Vote2026.06 | 75.28 | — | |
| Base Judge2026.06 | 74.49 | 23.87 | |
| Monolingual Vote2026.06 | 74.42 | 23.79 |
| Method | Links | ||
|---|---|---|---|
| SEMJ2026.06 | 78.84 | 32.07 | |
| Monolingual Ref2026.06 | 75.56 | 21 | |
| Multilingual Vote2026.06 | 75.28 | — | |
| Base Judge2026.06 | 74.49 | 23.87 | |
| Monolingual Vote2026.06 | 74.42 | 23.79 |