List your company

Embodied AI Task Execution on ALFWorld v1 (ID)

100Success Rate (%)AgentMark
76.70482.75288.894.848Jan 5, 2026

Evaluation results

MethodLinks
AgentMarkTask=A2: Pick And Place Simple2026.01100111.06140.290.25
BaseTask=A1: Look At Obj In Light2026.0197.112.8————
AgentMarkTask=A1: Look At Obj In Light2026.0197.111.51.1914.20.44-0.2
RGTask=A1: Look At Obj In Light2026.0196.213.8————
AgentMarkTask=A4: Pick Cool Then Place2026.019117.21.1726.4-0.140.42
AgentMarkTask=Avg.2026.0189.319.41.1925.50.10.2
AgentMarkTask=A5: Pick Heat Then Place2026.0188.919.41.3734.30.26-0.08
AgentMarkTask=A3: Pick Clean Then Place2026.0187.318.61.2330.70.050.12
AgentMarkTask=A6: Pick Two Obj And Place2026.0177.631.31.2837.1-0.130.19