List your company

Uncertainty Calibration on GPQA main

0.672AUROCPE
0.518080.558040.5980.63796Jan 2, 2026

Evaluation results

MethodLinks
PEBackbone=Qwen2.5-32B-Instruct2026.010.672
p(True)Backbone=Qwen2.5-7B-Instruct2026.010.649
ET-PEBackbone=Qwen2.5-32B-Instruct2026.010.636
PEBackbone=Qwen2.5-14B-Instruct2026.010.635
LSBackbone=Qwen2.5-32B-Instruct2026.010.634
ET-PEBackbone=Qwen2.5-14B-Instruct2026.010.622
LSBackbone=Qwen2.5-14B-Instruct2026.010.613
LSBackbone=Qwen2.5-7B-Instruct2026.010.601
p(True)Backbone=Qwen2.5-32B-Instruct2026.010.591
ET-PEBackbone=Qwen2.5-7B-Instruct2026.010.59
PEBackbone=Qwen2.5-7B-Instruct2026.010.586
p(True)Backbone=Qwen2.5-14B-Instruct2026.010.586
SEBackbone=Qwen2.5-7B-Instruct2026.010.544
LN-PEBackbone=Qwen2.5-7B-Instruct2026.010.541
SEBackbone=Qwen2.5-14B-Instruct2026.010.532
SEBackbone=Qwen2.5-32B-Instruct2026.010.528
LN-PEBackbone=Qwen2.5-14B-Instruct2026.010.526
LN-PEBackbone=Qwen2.5-32B-Instruct2026.010.524