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

Mathematical Reasoning on Source domain

65.9Accuracy

LALoRA

59.1460.89562.6564.405Dec 19, 2025
Updated 1mo ago

Evaluation Results

MethodLinks
2025.12
65.95
2025.12
65.74.9
2025.12
65.74.8
2025.12
65.74.8
2025.12
65.24.3
2025.12
64.94
2025.12
64.53.7
2025.12
643.1
2025.12
63.82.9
2025.12
63.82.9
2025.12
63.62.7
2025.12
63.52.7
2025.12
63.32.4
2025.12
63.22.4
2025.12
62.71.9
2025.12
62.71.8
2025.12
62.71.8
2025.12
62.51.6
2025.12
62.41.5
2025.12
62.31.4
2025.12
62.21.4
2025.12
61.70.9
2025.12
61.60.7
2025.12
610.1
2025.12
60.90
2025.12
60.8-0.1
2025.12
60.8-0.1
2025.12
60.80
2025.12
60.7-0.2
2025.12
59.4-1.4