Uncertainty Quantification on MATH
81.7AUROCP(true)
Evaluation Results
| Method | Links | |
|---|---|---|
| P(true)Model=Gemma2026.06 | 81.7 | |
| LogTokUModel=Gemma2026.06 | 80.1 | |
| Add-S˜Model=Gemma2026.06 | 80.1 | |
| GLU-AUModel=Fanar2026.06 | 80.1 | |
| GLUModel=Gemma2026.06 | 79.9 | |
| GLUModel=Fanar2026.06 | 78.8 | |
| RAUQModel=Fanar2026.06 | 77.9 | |
| GLU-Sα-SPModel=Qwen2026.06 | 77.2 | |
| RAUQModel=Gemma2026.06 | 77.1 | |
| GLUModel=Qwen2026.06 | 76.7 | |
| P(true)Model=Qwen2026.06 | 75.9 | |
| LogProbModel=Fanar2026.06 | 74.6 | |
| RAUQModel=Qwen2026.06 | 69.9 | |
| LogProbModel=Qwen2026.06 | 68.4 | |
| LogProbModel=Gemma2026.06 | 68.3 | |
| P(true)Model=Fanar2026.06 | 66.6 | |
| LogTokUModel=Fanar2026.06 | 66.4 | |
| LogTokUModel=Qwen2026.06 | 59.5 |