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

Full Set

Benchmarks

Task NameDataset NameSOTA ResultTrend
Tabular Classification and RegressionFull set of 158 datasets CARTE TabArena-Lite TALENT-Tiny TextTab
Mean Per-Task Rank2.7
26
Image Manipulation LocalizationFull Set (test)
mIoU78.9
8
Human-AI HybridizationFull Set 60% held-out N=952 (test)
Majority Accuracy69.3
5
Showing 3 of 3 rows