Natural Language Understanding on Polish Open Leaderboard (test)
70.36Polemo2-in AccQwen3
Evaluation results
| Method | Links | |||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Qwen3format=multiple-choice, shot=0-shot, template=chat template2025.12 | 70.36 | 59.92 | 63.16 | 56.67 | 13.96 | 30.97 | 8.75 | 50.04 | 29.2 | 48.6 | 43.16 | |
| Gamayunformat=multiple-choice, shot=0-shot, template=chat template2025.12 | 67.87 | 63.16 | 59.68 | 62.44 | 15.03 | 43.97 | 36.35 | 68.01 | 47.9 | 67.42 | 53.18 | |
| Gemma3format=multiple-choice, shot=0-shot, template=chat template2025.12 | 67.31 | 58.31 | 54.3 | 37.44 | 13.47 | 34.74 | 27.41 | 62.83 | 39.2 | 72.65 | 46.76 | |
| Llama3.2format=multiple-choice, shot=0-shot, template=chat template2025.12 | 46.81 | 34.82 | 17.82 | 36.33 | 14.93 | 29.23 | 24.25 | 52.66 | 39.6 | 46.66 | 34.31 | |
| EuroLMformat=multiple-choice, shot=0-shot, template=chat template2025.12 | 45.98 | 38.06 | 42.61 | 23.33 | 2.69 | 29 | 6.41 | 51.73 | 13.5 | 46.66 | 33.3 | |
| Qwen2.5format=multiple-choice, shot=0-shot, template=chat template2025.12 | 35.6 | 41.9 | 46.25 | 60.67 | 3.33 | 33.73 | 29.49 | 67.09 | 41.4 | 43.9 | 40.33 |