List your company

Probabilistic Model Calibration on TriviaQA, CoQA, and NQ

0.03ECEACSE
0.02440.06220.10.1378May 5, 2026

Evaluation results

MethodLinks
ACSEalpha (α)=0.20, Tolerance level=Permissive2026.050.030.08
ACSEalpha (α)=0.10, Tolerance level=Standard2026.050.040.09
CAPalpha (α)=0.20, Tolerance level=Permissive2026.050.050.1
CAPalpha (α)=0.10, Tolerance level=Standard2026.050.060.11
ACSEalpha (α)=0.05, Tolerance level=Strict2026.050.060.1
DDCRP-CPalpha (α)=0.20, Tolerance level=Permissive2026.050.070.12
CAPalpha (α)=0.05, Tolerance level=Strict2026.050.070.12
DDCRP-CPalpha (α)=0.10, Tolerance level=Standard2026.050.090.14
DDCRP-CPalpha (α)=0.05, Tolerance level=Strict2026.050.10.15
SUalpha (α)=0.20, Tolerance level=Permissive2026.050.140.22
SUalpha (α)=0.10, Tolerance level=Standard2026.050.160.24
SUalpha (α)=0.05, Tolerance level=Strict2026.050.170.25