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

SIVAL-MIPL

Benchmarks

Task NameDataset NameSOTA ResultTrend
ClassificationSIVAL-MIPL r=3 (test)
Accuracy69.13
12
ClassificationSIVAL-MIPL (r=2) (test)
Accuracy71.93
12
ClassificationSIVAL-MIPL (r=1) (test)
Accuracy77.65
12
Expected Calibration ErrorSIVAL-MIPL r=3 (test)
Reduction in Error45.47
6
Expected Calibration ErrorSIVAL-MIPL r=2 (test)
Reduction in Error47.71
6
Expected Calibration ErrorSIVAL-MIPL r=1 (test)
Reduction in Error49.52
6
Showing 6 of 6 rows