List your company

Mathematical Reasoning on OlympiadBench (ECE, Brier score)

7ECEProbe
5.85813.566521.27528.9835Mar 3, 2026

Evaluation results

MethodLinks
ProbeN=12026.03725.43
MSV 64N=642026.038.0326.13
MSV 1N=12026.038.0626.2
MSV 4N=42026.0310.3128.64
MSV 16N=162026.0310.5928.59
ProbeN=42026.0315.133.76
MSV 1N=42026.0316.7536.24
Probe + WV 64N=642026.0317.3641.02
MSV 1 + WV 64N=642026.0317.6942.19
MSV 1 + WV 16N=162026.0320.4944.76
Probe + WV 16N=162026.0320.7344.05
ProbeN=162026.0322.9444.28
MSV 1N=162026.0325.1948.3
MSV 1 + WV 4N=42026.0330.4456.42
ProbeN=642026.0330.4855.2
Probe + WV 4N=42026.0330.6656.6
MSV 1N=642026.0335.5564.53