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

HI

Benchmarks

Task NameDataset NameSOTA ResultTrend
ClassificationHI
Accuracy0.821
59
ClassificationHI 50% corrupted features
Accuracy82.1
21
Feature SelectionHI
Precision0.861
17
Feature SelectionHI with random extraneous features
ROC AUC92.2
17
Money Laundering Risk RankingHI Medium
AUC0.7699
16
Money Laundering Risk RankingHI Small
AUC78.79
16
ClassificationHI
Accuracy81.9
14
ClassificationHI corrupted features
Accuracy81.9
14
ClassificationHI 75% corrupted features (75% noise)
Accuracy (75% Noise)81.3
10
Federated Learning TrainingHI-Large
Average Runtime (s)86.68
8
Federated Learning TrainingHI Medium
Average Runtime (s)21.12
8
Federated Learning TrainingHI-Small
Average Runtime (s)18.79
8
Spell CorrectionHI Hindi
Det Rate89.2
6
Group recoveryHI Medium
Precision35.1
5
Group recoveryHI Small
Precision30.3
5
ClassificationHI
AUC0.887
2
Synthetic Tabular Data GenerationHI
DCR48.3
2
Showing 17 of 17 rows