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

RoboFail

Benchmarks

Task NameDataset NameSOTA ResultTrend
Failure DetectionRoboFail
Accuracy67
11
Failure DetectionRoboFail (Out-Of-Domain)
Execution Binary Acc86
8
Failure Taxonomy RecoveryRoboFail expert taxonomy
CP92
4
Robot Failure ExplanationRoboFail
Coherence Score (CS)0.628
3
Showing 4 of 4 rows