secret-word task

42.4Accuracy

Bootstrap

Updated 8d ago

Evaluation Results

Method
Bootstrap 2026.05	42.4	19.3	21.3	1.029	81
steering-coefficient sensitivity 2026.05	41.8	40.4	35.4	4.401	76.3
MCMC power-sampling acceptance 2026.05	41.5	54.4	53.4	10.77	56.3
Log-prob 2026.05	41.4	25.5	24.6	0.73	84
Bootstrap 2026.05	41.4	9.7	17.1	0.558	83
Direct (numeric) 2026.05	41.4	58.2	58	12.66	51.6
MCMC power-sampling agreement 2026.05	41.3	26.3	24.7	1.608	80.3
Bootstrap 2026.05	40.2	5.7	16.3	0.498	82.9
Bootstrap 2026.05	38.7	7.6	17.1	0.519	82.3
Bootstrap 2026.05	36.7	8.3	17.3	0.522	82.4
Log-prob 2026.05	23.4	13.1	15.8	0.489	83.5
MCMC power-sampling acceptance 2026.05	23.4	74	72.9	15.61	52.8
Direct (numeric) 2026.05	23.4	75.3	75.2	16.02	40.4
steering-coefficient sensitivity 2026.05	23.2	50.1	41.3	4.172	72.5
Bootstrap 2026.05	23.1	32.7	22.2	0.8	86.4
Bootstrap 2026.05	23	21.8	15.7	0.514	86.3
MCMC power-sampling agreement 2026.05	22.8	39	27.4	1.383	85.8
Bootstrap 2026.05	22.5	14.7	13	0.431	85.1
Bootstrap 2026.05	20.9	12.5	12.5	0.415	83.7
Bootstrap 2026.05	20.6	10.3	12.9	0.422	81.2