List your company

secret-word task

42.4AccuracyBootstrap
19.72825.61431.537.386May 25, 2026

Evaluation results

MethodLinks
BootstrapModel=Qwen3-8B, Temperature=0.52026.0542.419.321.31.02981
steering-coefficient sensitivityModel=Qwen3-8B2026.0541.840.435.44.40176.3
MCMC power-sampling acceptanceModel=Qwen3-8B, Temperature=0.1252026.0541.554.453.410.7756.3
Log-probModel=Qwen3-8B, Variant=with offset, baseline2026.0541.425.524.60.7384
BootstrapModel=Qwen3-8B, Temperature=0.72026.0541.49.717.10.55883
Direct (numeric)Model=Qwen3-8B2026.0541.458.25812.6651.6
MCMC power-sampling agreementModel=Qwen3-8B, Temperature=0.52026.0541.326.324.71.60880.3
BootstrapModel=Qwen3-8B, Temperature=1.02026.0540.25.716.30.49882.9
BootstrapModel=Qwen3-8B, Temperature=1.32026.0538.77.617.10.51982.3
BootstrapModel=Qwen3-8B, Temperature=1.52026.0536.78.317.30.52282.4
Log-probModel=Qwen3.6-27B, Variant=with offset, baseline2026.0523.413.115.80.48983.5
MCMC power-sampling acceptanceModel=Qwen3.6-27B, Temperature=0.1252026.0523.47472.915.6152.8
Direct (numeric)Model=Qwen3.6-27B2026.0523.475.375.216.0240.4
steering-coefficient sensitivityModel=Qwen3.6-27B2026.0523.250.141.34.17272.5
BootstrapModel=Qwen3.6-27B, Temperature=0.52026.0523.132.722.20.886.4
BootstrapModel=Qwen3.6-27B, Temperature=0.72026.052321.815.70.51486.3
MCMC power-sampling agreementModel=Qwen3.6-27B, Temperature=0.52026.0522.83927.41.38385.8
BootstrapModel=Qwen3.6-27B, Temperature=1.02026.0522.514.7130.43185.1
BootstrapModel=Qwen3.6-27B, Temperature=1.32026.0520.912.512.50.41583.7
BootstrapModel=Qwen3.6-27B, Temperature=1.52026.0520.610.312.90.42281.2