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

Arithmetic Reasoning on Countdown (Accuracy)

33.6Accuracy

LIFT2

-1.16727.858916.88525.9111Sep 25, 2025Nov 3, 2025Dec 13, 2025Jan 22, 2026Mar 2, 2026Apr 11, 2026May 21, 2026
Updated 9d ago

Evaluation Results

MethodLinks
2026.05
33.6
2025.09
27.5
2025.09
26
2026.05
25.6
2026.05
25
2026.05
23.4
2025.09
23.2
2026.05
22.3
2025.09
21.7
2026.05
21.1
2025.09
0.281
2025.09
0.218
2025.09
0.213
2025.09
0.211
2025.09
0.207
2025.09
0.188
2025.09
0.182
2025.09
0.173
2025.09
0.17