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

Robustness

Benchmarks

Task NameDataset NameSOTA ResultTrend
Image ClassificationRobustness (test)
Accuracy64.1
11
RobustnessRobustness
RR87.97
9
Text-to-SQL generationRobustness perturbations
Mean Robustness Weighted Average Score85.9
6
Watermarking RobustnessRobustness (test)
BER (40–49 cm)0.78
6
Lateral Tracking ControlRobustness Case 4 (test)
Lateral Error RMS0.031
3
Lateral Tracking ControlRobustness Case 3 (test)
Lateral Error RMS0.0311
3
Lateral Tracking ControlRobustness Case 2 (test)
Lateral Error RMS0.0341
3
Lateral Tracking ControlRobustness Case 1 (test)
Lateral Error RMS0.0341
3
Showing 8 of 8 rows