Multilingual Evaluation on SIB200
88.9HLFRMACRO
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| MACROModel=Gemma3-12B, Language=ar2026.05 | 88.9 | 40.7 | |
| MACROModel=Gemma3-12B, Language=zh2026.05 | 86.9 | 70.9 | |
| DG-CFModel=Gemma3-12B, Language=zh2026.05 | 78.9 | 55.4 | |
| DG-CFModel=Gemma3-12B, Language=ar2026.05 | 78.9 | 46.7 | |
| MACROModel=Gemma3-12B, Language=en2026.05 | 78.8 | 37.7 | |
| MACROModel=Gemma3-12B, Language=de2026.05 | 78.4 | 35.2 | |
| MACROModel=Gemma3-4B, Language=zh2026.05 | 77.8 | 66.6 | |
| DG-CFModel=Qwen3-4B, Language=zh2026.05 | 76.5 | 45.9 | |
| DG-CFModel=Gemma3-12B, Language=en2026.05 | 76.3 | 47.1 | |
| MACROModel=Gemma3-12B, Language=sw2026.05 | 74.7 | 35.8 | |
| MACROModel=Gemma3-12B, Language=vi2026.05 | 73.7 | 40.1 | |
| DG-CFModel=Gemma3-12B, Language=sw2026.05 | 73.3 | 44.6 | |
| MACROModel=Gemma3-12B, Language=ru2026.05 | 71.7 | 37.8 | |
| DG-CFModel=Gemma3-12B, Language=vi2026.05 | 70.9 | 50.8 | |
| DG-CFModel=Gemma3-12B, Language=de2026.05 | 70.3 | 47.2 | |
| DG-CFModel=Gemma3-12B, Language=ru2026.05 | 69.1 | 49.4 | |
| MACROModel=Qwen3-4B, Language=zh2026.05 | 66.2 | 57.4 | |
| MACROModel=Gemma3-4B, Language=ar2026.05 | 64.6 | 39.9 | |
| MACROModel=Qwen3-8B, Language=ar2026.05 | 64.6 | 28.1 | |
| MACROModel=Qwen3-4B, Language=vi2026.05 | 63.7 | 33.7 | |
| DG-CFModel=Gemma3-4B, Language=zh2026.05 | 62.9 | 46.7 | |
| DG-CFModel=Qwen3-4B, Language=vi2026.05 | 62.7 | 31.8 | |
| MACROModel=Qwen3-8B, Language=zh2026.05 | 62.6 | 56.6 | |
| MACROModel=Qwen3-4B, Language=en2026.05 | 61.8 | 29.4 | |
| DG-CFModel=Qwen3-8B, Language=vi2026.05 | 61.4 | 30.8 | |
| MACROModel=Gemma3-4B, Language=sw2026.05 | 60.6 | 47.3 | |
| MACROModel=Gemma3-4B, Language=ru2026.05 | 60.6 | 43.5 | |
| MACROModel=Gemma3-4B, Language=vi2026.05 | 58.6 | 43.7 | |
| MACROModel=Qwen3-8B, Language=de2026.05 | 58.6 | 25.9 | |
| MACROModel=Qwen3-4B, Language=de2026.05 | 57.8 | 26.3 | |
| MACROModel=Gemma3-4B, Language=en2026.05 | 57.6 | 42.7 | |
| MACROModel=Qwen3-4B, Language=ar2026.05 | 57.4 | 31 | |
| DG-CFModel=Qwen3-8B, Language=ru2026.05 | 57.2 | 29.5 | |
| DG-CFModel=Qwen3-4B, Language=ar2026.05 | 56.9 | 25.5 | |
| DG-CFModel=Qwen3-8B, Language=de2026.05 | 56.8 | 26.3 | |
| MACROModel=Qwen3-4B, Language=ru2026.05 | 56.4 | 27.3 | |
| DG-CFModel=Gemma3-4B, Language=en2026.05 | 55.9 | 52.9 | |
| MACROModel=Qwen3-8B, Language=en2026.05 | 55.6 | 30.8 | |
| MACROModel=Qwen3-8B, Language=vi2026.05 | 55.6 | 30.4 | |
| DG-CFModel=Gemma3-4B, Language=sw2026.05 | 54.9 | 44.6 | |
| DG-CFModel=Gemma3-4B, Language=ar2026.05 | 54.8 | 40.1 | |
| MACROModel=Gemma3-4B, Language=de2026.05 | 54.5 | 35.5 | |
| DG-CFModel=Qwen3-4B, Language=de2026.05 | 54.4 | 27.6 | |
| DG-CFModel=Qwen3-8B, Language=zh2026.05 | 53.4 | 32 | |
| DG-CFModel=Gemma3-4B, Language=ru2026.05 | 51.4 | 45.9 | |
| DG-CFModel=Gemma3-4B, Language=vi2026.05 | 50.8 | 45.6 | |
| DG-CFModel=Qwen3-8B, Language=ar2026.05 | 50.5 | 16.9 | |
| MACROModel=Qwen3-8B, Language=ru2026.05 | 50.5 | 28.3 | |
| DG-CFModel=Gemma3-4B, Language=de2026.05 | 49.9 | 38.9 | |
| DG-CFModel=Qwen3-8B, Language=en2026.05 | 48.9 | 23.1 | |
| DG-CFModel=Qwen3-4B, Language=en2026.05 | 48.5 | 25.1 | |
| DG-CFModel=Qwen3-4B, Language=ru2026.05 | 37.3 | 18.5 | |
| MACROModel=Qwen3-8B, Language=sw2026.05 | 26.3 | 75.5 | |
| DG-CFModel=Qwen3-8B, Language=sw2026.05 | 24.4 | 32.4 | |
| MACROModel=Qwen3-4B, Language=sw2026.05 | 13.7 | 24.2 | |
| DG-CFModel=Qwen3-4B, Language=sw2026.05 | 12.7 | 43.4 |