List your company

Embodied Task Execution on VLABench Commonsense

57.3Progress Score (PS)π0.5
37.0242.28547.5552.815Jun 2, 2026

Evaluation results

MethodLinks
π0.5mode=Zero-shot2026.0657.343.960
ERVLAmode=Zero-shot2026.06554457.2
Choice + No Knowledge TruncationAblation=Choice without KT, mode=Zero-shot2026.0652.64356.4
No Choice + Knowledge InsulationAblation=No choice and KI, mode=Zero-shot2026.065041.255
π0-FASTmode=Zero-shot2026.0648.63856.8
X-VLAmode=Zero-shot2026.0648.2——
No CoTAblation=No explicit chain-of-thought, mode=Zero-shot2026.0647.838.854
No Choice (End-to-End)Ablation=End-to-End No Choice, mode=Zero-shot2026.0645.436.452.8
π0mode=Zero-shot2026.064329.154.9
ACoT-VLAmode=Zero-shot2026.0637.8—52.3