Multitask Language Understanding on MMLU (Per-Language Accuracy Breakdown)
89Arabic Accuracy
o1
Updated 1mo ago
Evaluation Results
| Method | Links | |||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
2024.12 | 89 | 87.34 | 88.92 | 92.3 | 89.32 | 89.04 | 88.33 | 88.61 | 89.7 | 88.87 | 88.24 | 89.52 | 89.92 | 85.4 | 75.38 | |
2024.12 | 88.21 | 86.22 | 88 | 90.8 | 88.61 | 85.73 | 87.82 | 88.21 | 88.72 | 87.88 | 88.15 | 88.59 | 88.93 | 84.79 | 73.73 | |
2024.12 | 81.55 | 80.07 | 83.35 | 88.7 | 84.37 | 82.92 | 80.61 | 83.44 | 84.35 | 82.87 | 82.62 | 84.27 | 84.93 | 77.08 | 61.95 | |
2024.12 | 79.45 | 77.25 | 81.8 | 85.2 | 82.12 | 81.22 | 78.87 | 81.74 | 82.22 | 81.29 | 80.2 | 82.43 | 83.03 | 70.15 | 58.07 | |
2024.12 | 70.89 | 65.77 | 73.05 | 82 | 76.59 | 74.31 | 69.16 | 74.52 | 76.4 | 72.55 | 72.03 | 76.77 | 77.37 | 61.91 | 45.83 |