Multilingual LLM Evaluation on Multilingual Benchmarks 18 Languages
62.78Reading ComprehensionMuRater
Evaluation Results
| Method | Links | ||||
|---|---|---|---|---|---|
| MuRaterModel Size=7B2025.07 | 62.78 | 64.4 | 44.5 | 57.23 | |
| QuRater-MModel Size=7B2025.07 | 61.96 | 63.28 | 43.31 | 56.18 | |
| MuRater(E)Language Setup=18 Languages2025.07 | 56.05 | 56.42 | 40.4 | 50.96 | |
| MuRater(M)Language Setup=18 Languages2025.07 | 54.91 | 55.48 | 39.68 | 50.02 | |
| QuRater-MLanguage Setup=18 Languages2025.07 | 54.58 | 54.87 | 38.12 | 49.19 | |
| UniformLanguage Setup=18 Languages2025.07 | 53.16 | 54.58 | 38.25 | 48.66 | |
| FineWeb-2Language Setup=18 Languages2025.07 | 50.83 | 52.48 | 35.53 | 46.28 | |
| HPLT-2Language Setup=18 Languages2025.07 | 50.38 | 49.77 | 36.96 | 45.7 |