Share your thoughts, 1 month free Claude Pro on usSee more

Scientific Problem Solving on FrontierScience-Olympiad

5.59Token Efficiency Ratio (B_method/BMV)

ESC sweep

Updated 23d ago

Evaluation Results

Method	Links
ESC sweep 2026.05		5.59	33.8	50	107
PC-linear 2026.05		3.88	33.8	50	2.5
ESC sweep 2026.05		2.74	33.8	50	45
AC sweep 2026.05		1.93	33.8	50	45
AC sweep 2026.05		1.54	33.8	50	107
P(True) 2026.05		1.23	33.8	50	45
P(True) 2026.05		1.2	33.8	50	107
Self-certainty 2026.05		1.04	29.5	49.3	486
Self-certainty 2026.05		1.01	29.5	49.3	1.3
ESC sweep 2026.05		0.94	33.8	50	2.5
DeepConf tail 2026.05		0.91	33.8	50	45
DeepConf tail 2026.05		0.91	33.8	50	107
Self-certainty 2026.05		0.89	33.8	50	45
Self-certainty 2026.05		0.87	33.8	50	107
Self-certainty 2026.05		0.85	29.5	49.3	6
PC-linear 2026.05		0.7	33.8	50	45
PC-quadratic 2026.05		0.65	33.8	50	45
PC-cubic 2026.05		0.65	33.8	50	45
PC-linear 2026.05		0.63	33.8	50	107
PC-quadratic 2026.05		0.52	33.8	50	107
PC-cubic 2026.05		0.52	33.8	50	107
AC sweep 2026.05		0.29	33.8	50	2.5
P(True) 2026.05		0.23	33.8	50	2.5
PC-cubic 2026.05		0.22	29.5	49.3	6
PC-linear 2026.05		0.1	9.1	14.2	924
PC-quadratic 2026.05		0.06	33.8	50	2.5
PC-cubic 2026.05		0.05	33.8	50	2.5