Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

Agentic Reasoning on ALFWorld (Action Success Rates)

97.2Pick Success Rate

HölderPO

84.82488.03791.2594.463May 12, 2026
Updated 21d ago

Evaluation Results

MethodLinks
2026.05
97.285.787.591.779.281.587.5
2026.05
96.910010010085.784.593.8
2026.05
94.467.594.894.479.876.486.7
2026.05
93.178.68188.282.189.585.9
2026.05
85.353.784.578.259.753.572.8