Natural Questions On Theoremqa
評価指標
Accuracy
評価結果
このベンチマークにおける各モデルのパフォーマンス結果
比較表
| モデル名 | Accuracy |
|---|---|
| theoremqa-a-theorem-driven-question-answering | 52.4 |
| theoremqa-a-theorem-driven-question-answering | 43.8 |
| dart-math-difficulty-aware-rejection-tuning-1 | 15.4 |
| theoremqa-a-theorem-driven-question-answering | 35.6 |
| dart-math-difficulty-aware-rejection-tuning-1 | 27.4 |
| theoremqa-a-theorem-driven-question-answering | 24.9 |
| theoremqa-a-theorem-driven-question-answering | 25.9 |
| dart-math-difficulty-aware-rejection-tuning-1 | 16.4 |
| dart-math-difficulty-aware-rejection-tuning-1 | 28.2 |
| dart-math-difficulty-aware-rejection-tuning-1 | 32.2 |
| theoremqa-a-theorem-driven-question-answering | 30.2 |
| theoremqa-a-theorem-driven-question-answering | 23.6 |
| theoremqa-a-theorem-driven-question-answering | 22.8 |
| theoremqa-a-theorem-driven-question-answering | 21.0 |
| dart-math-difficulty-aware-rejection-tuning-1 | 19.4 |
| dart-math-difficulty-aware-rejection-tuning-1 | 17.0 |
| theoremqa-a-theorem-driven-question-answering | 23.9 |
| dart-math-difficulty-aware-rejection-tuning-1 | 32.5 |
| theoremqa-a-theorem-driven-question-answering | 31.8 |