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

Civil

Benchmarks

Task NameDataset NameSOTA ResultTrend
Toxicity DetectionCivil
Accuracy87.06
10
Zero-shot forecasting on multivariate time seriesCivil Generative Avg.
MAE0.525
5
Toxicity ClassificationCivil reconstructed with controlled shortcut injection (test)
MSTPS54.9
5
Showing 3 of 3 rows