Question Answering on TruthfulQA (Accuracy and Utility)
68.2Accuracy
Mistral-v2-7B
Updated 8d ago
Evaluation Results
| Method | Links | ||
|---|---|---|---|
2026.05 | 68.2 | 56.4 | |
2026.05 | 66.9 | -26 | |
2026.05 | 66.6 | 31.4 | |
| 66.6 | 36.6 | ||
| 66.5 | 24.5 | ||
2026.05 | 65.9 | 20.7 | |
2026.05 | 62.4 | 74.8 | |
| 60 | 53.5 | ||
| 59.7 | 46.8 | ||
2026.05 | 59.5 | 49.8 | |
2026.05 | 59.2 | 40.3 | |
2026.05 | 57.3 | 42.2 | |
2026.05 | 51.6 | 56.5 | |
2026.05 | 51.3 | 14 | |
| 51.2 | 46.5 | ||
| 51 | 23.5 | ||
2026.05 | 51 | 37.9 | |
2026.05 | 50.4 | 30.1 |