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

Question Answering Calibration on HotpotQA (test)

65.7Accuracy

LIMECAL

Updated 3mo ago

Evaluation Results

Method	Links
LIMECAL 2021.10		65.7	78.2	82.6	78.4	73.8	-
SHAPCAL 2021.10		65.3	77.8	82	78	73.5	-
BOWPROP 2021.10		64.7	76.6	80.3	76.9	72.4	0
KAMATH 2021.10		64.5	76.8	80.8	77.2	72.8	-
MAXPROB 2021.10		63.1	75.7	79.7	75.9	72.2	-