ResearchBenchmarksControl on OGBench-CubeFollow93.7Success RateQantara63.85271.60179.3587.099Jul 6, 2026Evaluation ResultsMethodMethodLinksSuccess RateQantaraEvaluation pipeline=Cu...Evaluation pipeline=Current paper reproduction2026.0793.7DINO-WMEvaluation pipeline=Ma...Evaluation pipeline=Maes et al. [29]2026.0786LeWMEvaluation pipeline=Cu...Evaluation pipeline=Current paper reproduction2026.0776.7LeWMEvaluation pipeline=Ma...Evaluation pipeline=Maes et al. [29]2026.0774PLDMEvaluation pipeline=Ma...Evaluation pipeline=Maes et al. [29]2026.0765