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

PHACT Safety Suite

Benchmarks

Task NameDataset NameSOTA ResultTrend
Certified Task-solvingPHACT Safety Suite (Impossible goals)
False Certifications (Bare)1
2
Certified Task-solvingPHACT Safety Suite Feasible goals
Bare Feasible Valid Score52
2
Showing 2 of 2 rows