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

Normalized Accuracy on ARC-E (Question Answering)

79.5Normalized Accuracy (ARC-E)

Base

-14.30810.04634.458.754Apr 6, 2026Apr 7, 2026
Updated 10d ago

Evaluation Results

MethodLinks
2026.04
79.5
2026.04
79.5
2026.04
77.4
2026.04
76.4
2026.04
46.6
2026.04
42.8
2026.04
29.4
2026.04
29
2026.04
28.7
2026.04
3.8
2026.04
1.5
2026.04
1.5
2026.04
1.4
2026.04
0.5
2026.04
0.2
2026.04
-0.2
2026.04
-1.4
2026.04
-1.9
2026.04
-10.7