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

Confidence Calibration on TriviaQA (test)

0.065Expected Calibration Error

DINCO

0.059840.094670.12950.16433Sep 29, 2025
Updated 1mo ago

Evaluation Results

MethodLinks
2025.09
0.0650.1310.863
2025.09
0.1130.1640.792
2025.09
0.1290.190.741
2025.09
0.1530.1580.862
2025.09
0.1660.2130.707
2025.09
0.1860.2110.737
2025.09
0.1940.1690.88