Our new X account is live! Follow @wizwand_team for updates
WorkDL logo mark

Embodied decision-making on ALFWorld

82.84Success Rate

UMEM-Qwen3-4B

28.510442.615256.7270.8248Jan 22, 2026Jan 25, 2026Jan 28, 2026Feb 1, 2026Feb 4, 2026Feb 7, 2026Feb 11, 2026
Updated 4d ago

Evaluation Results

MethodLinks
2026.02
82.84---------84.2
2026.02
77.61---------81.34
2026.01
74.3------0.9680.9050.791-
2026.02
73.13---------79.6
2026.01
72.9------0.9580.8560.774-
2026.02
70.9---------83.71
2026.02
70.9---------78.86
2026.01
69.5------0.9480.8470.763-
2026.01
67.9------0.9250.820.721-
2026.01
66.4------0.9220.8310.718-
2026.01
65.7------0.9630.8410.807-
2026.02
64.18---------75.37
2026.01
63.6------0.9130.7830.667-
2026.02
61.94---------66.67
2026.02
61.94---------73.63
2026.02
61.19---------78.61
2026.02
60.45---------76.74
2026.02
59.7---------74.5
2026.02
58.96---------77.24
2026.02
56.72---------75.62
2026.02
55.22---------72.89
2026.02
53.73---------71.27
2026.02
52.24---------72.64
2026.02
50.75---------73.13
2026.02
47.01---------74.13
2026.02
46.27---------66.17
2026.02
44.78---------69.78
2026.02
44.78---------72.14
2026.02
41.04---------68.91
2026.02
40.3---------65.92
2026.02
30.6---------54.23
2026.01
-0.3060.2580.2860.2720.2550.236----
2026.01
-0.2790.2370.2340.2640.1990.219----
2026.01
-0.2640.2270.2230.2540.1850.218----
2026.01
-0.1850.2240.1770.2150.1780.206----
2026.01
-0.1090.220.160.2220.1040.217----
2026.01
-0.2050.2020.2070.2050.1310.191----
2026.01
-0.1740.2180.1620.1980.0930.176----