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

BT

Benchmarks

Task NameDataset NameSOTA ResultTrend
Lesion SegmentationBT
mDice84.5
18
Expected Calibration ErrorBT
ECE (15-bin)0.025
9
Preference ModelingBT
NLL1.08
8
Prompt OptimizationBT
Best Score62
8
Multiple choice question answeringBT (test)
Accuracy39.6
7
ClassificationBT UCI (train)
Accuracy79.59
6
Showing 6 of 6 rows