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

Multi-modal Mathematical Reasoning on FlowVerse raw

70.4Performance

MathFlow-P-7B

38.36846.6845563.316Mar 19, 2025
Updated 1mo ago

Evaluation Results

MethodLinks
2025.03
70.4
2025.03
69.6
2025.03
68.9
2025.03
68.1
2025.03
64
2025.03
63.8
2025.03
62.2
2025.03
60.1
2025.03
59.6
2025.03
59.3
2025.03
58.3
2025.03
56.2
2025.03
54.7
2025.03
54.3
2025.03
52.3
2025.03
48.9
2025.03
48.2
2025.03
47.1
2025.03
43.5
2025.03
39.6