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

Zero-shot Question Answering on ARC-E

79.38Zero-shot Accuracy

Dense

22.689637.407352.12566.8427Oct 2, 2025
Updated 1mo ago

Evaluation Results

MethodLinks
2025.10
79.38
2025.10
76.3
2025.10
76.3
2025.10
75.84
2025.10
75.84
2025.10
75.12
2025.10
74.96
2025.10
72.98
2025.10
72.93
2025.10
72.43
2025.10
70.67
2025.10
70.58
2025.10
69.4
2025.10
68.81
2025.10
62.42
2025.10
61.2
2025.10
61.07
2025.10
59.64
2025.10
56.14
2025.10
50.97
2025.10
49.58
2025.10
36.45
2025.10
36.15
2025.10
35.99
2025.10
33.8
2025.10
31.31
2025.10
28.79
2025.10
28.37
2025.10
26.94
2025.10
26.26
2025.10
26.01
2025.10
25.93
2025.10
25.76
2025.10
25.76
2025.10
25.63
2025.10
24.87