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

Question Answering on ARC Easy (Accuracy and Normalized Accuracy)

67.26Normalized Accuracy

Softmax

25.056836.013446.9757.9266Apr 29, 2025Jun 28, 2025Aug 27, 2025Oct 27, 2025Dec 26, 2025Feb 24, 2026Apr 26, 2026
Updated 21d ago

Evaluation Results

MethodLinks
2025.04
67.2672.22
2025.04
66.7571.93
2025.04
62.2169.07
2025.04
58.8467.05
2025.04
56.3660.94
2025.04
56.3160.31
2025.04
54.2959.26
2025.04
53.9659.93
2025.08
53.87-
2025.04
52.5757.95
2025.04
51.8156.65
2025.04
48.6552.78
2025.04
48.5753.83
2025.04
36.3637.54
2026.04
35.6938.93
2026.04
34.7637.46
2025.04
29.7629.88
2025.08
29.34-
2025.04
27.2727.02
2025.04
26.6826.81