List your company

ResearchFormerly WizWand

RISK-Bench

Benchmarks

TaskDatasetSOTA resultTrendResults

E-commerce risk management multi-step trajectory evaluation

RISK-Bench Multi-step

82.8Multi-step Success Rate

19

E-commerce risk management single-step trajectory evaluation

RISK-Bench Single-step

99.1Easy Score

19

GUI Task Completion

RISK-Bench (online)

88.7Completion Rate

14