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

Question Answering on ARC-c (Acc, Len)

96.82Accuracy

Zero-shot-EI

24.789643.489862.1980.8902Jun 9, 2025Jul 30, 2025Sep 19, 2025Nov 9, 2025Dec 30, 2025Feb 19, 2026Apr 11, 2026
Updated 4d ago

Evaluation Results

MethodLinks
2026.02
96.82-
2026.02
92-
2026.02
91.78-
2026.02
91.45-
2026.02
91.08-
2026.02
90.55-
2026.02
90.12-
2026.02
90.1-
2026.02
89.64-
2026.02
89.05-
2026.02
88.4-
2026.02
87.26-
2026.02
85.36-
2025.06
81.61,321
2025.06
80.8639
2026.02
80.67-
2025.06
80.41,168
2026.02
79.94-
2026.02
78.15-
2026.02
75.09-
2025.06
751,515
2026.02
74.23-
2026.02
73.92-
2026.02
73.29-
2026.02
72.92-
72.8980
2026.02
72.74-
2026.02
72.37-
72.3879
2026.02
68.34-
65.72,881
63.51,814
2025.06
50.3504
2026.04
45.48-
2026.04
44.8-
2026.04
44.53-
2026.04
42.91-
2026.04
42.06-
2026.04
41.55-
2026.04
41.38-
2026.04
41.38-
2026.04
40.44-
2026.04
39.33-
2026.04
39.16-
2026.04
37.63-
2026.04
37.12-
2026.04
36.43-
2026.04
34.22-
2026.04
31.74-
2026.04
28.49-
2026.04
27.9-
2026.04
27.73-
2026.04
27.56-
2026.04
27.56-