Factual Accuracy Estimation on Real-data Factual Mistral responses
51.05Original AccuracyLogistic Regression
Evaluation results
| Method | Links | ||||
|---|---|---|---|---|---|
| Logistic RegressionQuestion Type=Factual, Data=8260, Verification Budget=3%2026.06 | 51.05 | 56.11 | 5.06 | 248 | |
| Multilayer PerceptronQuestion Type=Factual, Data=8260, Verification Budget=3%2026.06 | 51.05 | 53.9 | 2.84 | 248 | |
| Random ForestQuestion Type=Factual, Data=8260, Verification Budget=3%2026.06 | 51.05 | 59.96 | 8.91 | 248 | |
| Label PropagationQuestion Type=Factual, Data=8260, Verification Budget=3%2026.06 | 51.05 | 50.87 | -0.19 | 248 | |
| nnPUQuestion Type=Factual, Data=8260, Verification Budget=3%2026.06 | 51.05 | 56.32 | 5.27 | 248 | |
| AURAQuestion Type=Factual, Data=8260, Verification Budget=3%2026.06 | 51.05 | 72.37 | 21.31 | 259 |