Mathematical Reasoning on MATH (ECE, Brier Score)
7.26ECEProbe
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| ProbeN=12026.03 | 7.26 | 26.93 | |
| MSV 1N=12026.03 | 8.49 | 29.2 | |
| MSV 4N=42026.03 | 11.33 | 30.1 | |
| MSV 16N=162026.03 | 11.6 | 29.05 | |
| ProbeN=42026.03 | 11.68 | 30.01 | |
| Probe + WV 64N=642026.03 | 12.59 | 33.05 | |
| MSV 1 + WV 64N=642026.03 | 13 | 33.29 | |
| MSV 1N=42026.03 | 13.16 | 33.16 | |
| MSV 64N=642026.03 | 13.23 | 29.74 | |
| Probe + WV 16N=162026.03 | 13.53 | 34.07 | |
| MSV 1 + WV 16N=162026.03 | 13.56 | 34.52 | |
| ProbeN=162026.03 | 15.59 | 33.97 | |
| MSV 1N=162026.03 | 17.4 | 38.18 | |
| Probe + WV 4N=42026.03 | 17.97 | 38.56 | |
| MSV 1 + WV 4N=42026.03 | 18.15 | 38.71 | |
| ProbeN=642026.03 | 18.48 | 38.02 | |
| MSV 1N=642026.03 | 21.28 | 44.19 |