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

Multi-turn Agent Task on ALFWorld

85.16Success Rate

ATOD

-2.595220.187442.9765.7526Jun 26, 2026
Updated 27d ago

Evaluation Results

MethodLinks
2026.06
85.1616.75
2026.06
82.8115.84
2026.06
80.4722.69
2026.06
80.4723.53
2026.06
80.4721.12
2026.06
79.6919.02
2026.06
77.3420.85
2026.06
76.5618.95
2026.06
76.5618.48
2026.06
76.5622.9
2026.06
76.5618.95
2026.06
76.5623.45
2026.06
75.7827.95
2026.06
74.2220.15
2026.06
72.6616.41
2026.06
71.8818.88
2026.06
39.8437.56
2026.06
31.2540.38
2026.06
30.4744.13
2026.06
28.1245.27
2026.06
24.2242.43
2026.06
10.1647.91
2026.06
0.7850