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

Question Answering on ARC (pass@1)

88.68pass@1

MCDIFFUSE

5.594427.164748.73570.3053Jan 12, 2026Jan 17, 2026Jan 22, 2026Jan 28, 2026Feb 2, 2026Feb 7, 2026Feb 13, 2026
Updated 1mo ago

Evaluation Results

MethodLinks
2026.02
88.68
2026.02
88.4
2026.02
87.98
2026.02
87.8
2026.02
87.09
2026.01
86.43
2026.01
85.49
2026.02
84.95
2026.01
84.73
2026.01
84.56
2026.01
84.39
2026.02
84.09
2026.02
83.32
2026.01
82.17
2026.01
81.14
2026.02
80.98
2026.01
80.63
2026.01
80.63
2026.01
79.35
2026.01
79.01
2026.01
74.06
2026.01
72.95
2026.01
70.31
2026.01
69.2
2026.01
67.32
2026.01
63.57
2026.01
57.51
2026.01
57.08
2026.01
29.69
2026.01
8.79