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

Agent Task on AlfWorld

86.7Success Rate

ResRL

4.85226.10147.3568.599Jun 13, 2025Aug 10, 2025Oct 7, 2025Dec 4, 2025Jan 31, 2026Mar 30, 2026May 28, 2026
Updated 5d ago

Evaluation Results

MethodLinks
2026.05
86.7-90.185.5988378.784.2-
2026.02
83.664.1-------
82.7--------
2026.02
80.759-------
2026.02
80.762.9-------
2026.05
80.4-92.36492.589.580.368.8-
2026.02
78.658.1-------
2026.05
78.5-92.975.274.886.373.765.3-
75.4--------
2026.05
74.8-88.843.788.170.377.756.8-
74.4--------
73.8--------
72.9--------
2026.05
71.2--------
69--------
2026.05
67.1--------
2026.02
66.450.7-------
2026.02
62.148.1-------
2026.05
60.3-92.863.362.16926.658.7-
2026.02
59.346.7-------
2026.02
57.140.9-------
2026.02
52.138.3-------
2026.02
50.729-------
2026.05
48-75.360.831.256.721.649.8-
2026.02
37.929-------
2025.06
36.9-------160
2025.06
35-------225
31.5--------
2026.05
31.2-48.535.434.313.218.217.6-
2026.02
25.721.8-------
2026.02
2518.6-------
2026.02
22.115.8-------
2026.02
16.412.8-------
2025.12
16--------
2026.02
15.79.8-------
2025.12
15--------
2026.05
14.8-33.421.619.36.92.83.2-
2025.12
12--------
2026.02
8.67.6-------
2025.12
8--------