ResearchBenchmarksControl on Reacher HardFollow86Success RateLeWM63.43269.29175.1581.009Jul 6, 2026Evaluation ResultsMethodMethodLinksSuccess RateLeWMEvaluation pipeline=Ma...Evaluation pipeline=Maes et al. [29]2026.0786QantaraEvaluation pipeline=Cu...Evaluation pipeline=Current paper reproduction2026.0780.9DINO-WMEvaluation pipeline=Ma...Evaluation pipeline=Maes et al. [29]2026.0779PLDMEvaluation pipeline=Ma...Evaluation pipeline=Maes et al. [29]2026.0778LeWMEvaluation pipeline=Cu...Evaluation pipeline=Current paper reproduction2026.0764.3