ResearchBenchmarksEmbodied Action Planning on Droid Real-world TasksFollow75AUTOLab ScoreGEM-4D47.9654.986269.02May 20, 2026Evaluation ResultsMethodMethodLinksAUTOLab ScoreCLVR ScoreRAIL ScoreGEM-4D2026.05758387Tesseract2026.05586559CogVideoX2026.05496439