Our new X account is live! Follow @wizwand_team for updates
WorkDL logo mark

OLID

Benchmarks

Task NameDataset NameSOTA ResultTrend
Offensive Language IdentificationOLID
Accuracy Drop-7
40
Offensive Language IdentificationOLID (test)
CACC87.33
33
Text ClassificationOLID (test)
Delta CACC1.79
18
Backdoor DefenseOLID (test)
ΔCACC2.21
12
Backdoor Trigger DetectionOLID
Precision0.43
10
Offensive language identificationOLID English Sub-task A 1.0 (test)
Macro F10.735
9
Offensive language target identificationOLID English 2019 (test)
Macro F155.7
5
Categorization of offensive language typeOLID English Sub-task B
Macro F10.619
5
Safety EvaluationOLID
F1 Score73
3
Text Anomaly DetectionOLID
AUPRC0.1581
2
Text ClassificationOLID
Delta ASR-
0
Backdoor DefenseOLID
Delta CACC-
0
Showing 12 of 12 rows