ResearchBenchmarksInverse Reinforcement Learning on AntFollow91Normalized PerformanceTRIRL6.7628.6350.572.37May 10, 2026Evaluation ResultsMethodMethodLinksNormalized PerformanceTRIRLPhase=TrainingPhase=Training2026.0591TRIRLPhase=RetrainingPhase=Retraining2026.0563AIRLPhase=TrainingPhase=Training2026.0559NEARPhase=TrainingPhase=Training2026.0546NEARPhase=RetrainingPhase=Retraining2026.0546AIRLPhase=RetrainingPhase=Retraining2026.0510