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

Multimodal Mathematical Reasoning on FlowVerse Text Limited

40.5CoT Error

InternVL2.5-8B

39.16848.15957.1566.141Mar 19, 2025
Updated 1mo ago

Evaluation Results

MethodLinks
2025.03
40.538.4
2025.03
40.636.7
2025.03
44.437.4
2025.03
46.738.3
2025.03
49.942.9
2025.03
53.445.8
2025.03
54.345.7
2025.03
57.949.8
2025.03
58.253.2
2025.03
58.750.3
2025.03
58.754.4
2025.03
58.951.3
2025.03
60.852.2
2025.03
64.160.3
2025.03
6555
2025.03
66.160.8
2025.03
73.566.7
2025.03
73.867.2