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

Next-state Prediction Accuracy on ALFWorld (AW)

99.87EM Accuracy

Qwen2.5-7B

32.498849.989467.4884.9706Dec 21, 2025
Updated 1mo ago

Evaluation Results

MethodLinks
2025.12
99.87
2025.12
99.71
2025.12
77.04
2025.12
67.13
2025.12
64.73
2025.12
63.79
2025.12
63.37
2025.12
62.56
2025.12
61.85
2025.12
56.88
2025.12
50
2025.12
45.2
2025.12
44.45
2025.12
43.56
2025.12
42.64
2025.12
35.09