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

Mathematical Reasoning on MathVista (test)

70.52Accuracy

Qwen2.5-VL-7B + Cont. Reward

15.795230.002644.2158.4174Nov 6, 2023Mar 9, 2024Jul 11, 2024Nov 12, 2024Mar 16, 2025Jul 18, 2025Nov 20, 2025
Updated 1mo ago

Evaluation Results

MethodLinks
2025.11
70.52-----
2025.11
68.88-----
68.46-----
2025.11
68.43-----
2025.02
65.360.173.168.867.755.9
2025.02
63.864.7----
2025.02
60.348.47359.763.255.9
2024.05
56.5-----
2025.07
55.7-----
2025.07
55.3-----
2025.07
53.4-----
2024.05
53.3-----
2024.05
53-----
2025.02
52.544.759.755.452.549.7
2024.05
52.1-----
2025.02
51.166.854.839.457.640.8
2024.05
50.5-----
2024.05
49.9-----
2025.07
49.2-----
2025.07
48.8-----
2024.05
47.9-----
2025.07
47.8-----
2025.02
46.657.756.537.251.333.5
2025.02
46.53851.650.951.939.7
2024.05
46.4-----
2024.05
45.2-----
2025.02
43.335.531.254.648.151.4
2024.05
40-----
2024.05
37-----
2024.05
36.8-----
2025.02
36.716.423.154.641.843
2024.05
36.5-----
2024.05
35.9-----
2024.05
35.5-----
2024.05
35.3-----
2025.02
34.928.415.226.832.934.6
2023.11
34.5-----
2023.11
33.8-----
2024.05
33.5-----
2024.05
32.9-----
2024.05
32.7-----
2024.05
29.6-----
2024.05
28.6-----
2024.05
28.2-----
2023.11
27.8-----
2025.02
27.722.718.923.84330.2
2024.05
27.6-----
2023.11
26.1-----
2024.05
25.5-----
2023.11
25.3-----
2023.11
25.3-----
2025.02
25.148.73.619.12528.7
2023.11
23.6-----
2023.11
23.1-----
2025.02
23.12613.418.630.430.2
2024.05
22.2-----
2025.02
22.223.610.222.727.227.9
2023.11
18.6-----
2025.02
17.921.63.818.219.626.3