Math on Real-data Math GPT-5.4 outputs (3% verified data)
67.2Original AccuracyLogistic Regression
Evaluation Results
| Method | Links | ||||
|---|---|---|---|---|---|
| Logistic RegressionQuestion Type=Math, Data=28172026.06 | 67.2 | 61.05 | -6.15 | 85 | |
| Multilayer PerceptronQuestion Type=Math, Data=28172026.06 | 67.2 | 62 | -5.2 | 85 | |
| Random ForestQuestion Type=Math, Data=28172026.06 | 67.2 | 65.92 | -1.28 | 85 | |
| Label PropagationQuestion Type=Math, Data=28172026.06 | 67.2 | 67.07 | -0.13 | 85 | |
| nnPUQuestion Type=Math, Data=28172026.06 | 67.2 | 67.23 | 0.03 | 85 | |
| AURAQuestion Type=Math, Data=28172026.06 | 67.2 | 70.05 | 2.85 | 95.8 |