List your company

ResearchFormerly WizWand

Off-policy policy selection

Benchmarks

DatasetSOTA methodMetricTrendResultsLast updated

Chain environment 100 batches (Evaluation)

0.54Max ∆U

8

May 26, 2026

Ring environment 100 batches (Evaluation)

0.84Max Utility Difference

8

May 26, 2026