Uncertainty Quantification on MATH (PRR)
70PRRGLU-Sα-SP
Evaluation Results
| Method | Links | |
|---|---|---|
| GLU-Sα-SPModel=Gemma2026.06 | 70 | |
| GLUModel=Gemma2026.06 | 69.9 | |
| LogTokUModel=Gemma2026.06 | 69.3 | |
| GLU-AUModel=Fanar2026.06 | 64.8 | |
| RAUQModel=Gemma2026.06 | 63.9 | |
| P(true)Model=Gemma2026.06 | 63.6 | |
| GLUModel=Fanar2026.06 | 62.7 | |
| RAUQModel=Fanar2026.06 | 62.4 | |
| P(true)Model=Qwen2026.06 | 57.4 | |
| GLU-Sα-SPModel=Qwen2026.06 | 54.5 | |
| GLUModel=Qwen2026.06 | 53.7 | |
| LogProbModel=Fanar2026.06 | 52 | |
| LogProbModel=Gemma2026.06 | 43.8 | |
| RAUQModel=Qwen2026.06 | 42.4 | |
| LogProbModel=Qwen2026.06 | 38.1 | |
| P(true)Model=Fanar2026.06 | 32.5 | |
| LogTokUModel=Fanar2026.06 | 32.4 | |
| LogTokUModel=Qwen2026.06 | 21.6 |