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

Click

Benchmarks

Task NameDataset NameSOTA ResultTrend
Noisy label detectionClick
AUC0.855
18
Identifying mislabeled pointsclick
F1 Score9
12
Identifying mislabeled pointsclick
AUC-ROC11
12
Identifying mislabeled pointsclick
Precision7
12
Identifying mislabeled pointsclick
Recall14
12
ClassificationClick
Error Rate32.86
11
High-value data removalClick (test)
Weighted Accuracy Drop0.4
8
Cultural Question AnsweringCLIcK
Accuracy0.5888
7
KnowledgeCLIcK
Score80.9
7
Data Valuationclick
Valuation Runtime (s)0.31
5
Text-to-TextCLIcK Korean
Score75.2
4
Korean CultureCLICK
Score82.6
4
Verifiable Data Valuationclick
Proving time (s)11.3
3
Showing 13 of 13 rows