ResearchBenchmarksOff-policy prediction on Two-state (Steady-state AUC error)Follow1.05Steady-state AUC errorTDRC0.7322.87855.0257.1715May 17, 2026Evaluation ResultsMethodMethodLinksSteady-state AUC errorTDRC2026.051.05TD2026.052.718GTD2-MP2026.052.89GTD22026.053.14TDC2026.056.33BA-TDRC2026.059