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

Multimodal Science Question Answering on ScienceQA IMG

95.78Accuracy

FP16

55.968866.304476.6486.9756Apr 2, 2026Apr 3, 2026Apr 4, 2026Apr 5, 2026Apr 6, 2026Apr 7, 2026
Updated 11d ago

Evaluation Results

MethodLinks
2026.04
95.78-
2026.04
95.59-
2026.04
95.54-
2026.04
95.49-
2026.04
95.34-
2026.04
95.19-
2026.04
95.09-
2026.04
95.09-
2026.04
95.09-
2026.04
95.09-
2026.04
94.99-
2026.04
94.99-
2026.04
94.99-
2026.04
94.59-
2026.04
94.1-
2026.04
93.85-
2026.04
93.06-
2026.04
84.5376.53
2026.04
7755.64
2026.04
76.365.42
2026.04
73.5772.17
2026.04
73.4872.17
2026.04
73.4-
2026.04
73.2872.17
2026.04
73.2372.27
2026.04
72.9872.17
2026.04
72.8872.17
2026.04
72.8-
2026.04
72.7-
2026.04
72.7-
2026.04
72.5370.94
2026.04
72.5-
2026.04
72.4-
2026.04
72.3-
2026.04
72.2470.94
2026.04
72.2-
2026.04
72.1471.44
2026.04
72.1471.44
2026.04
72-
2026.04
71.9471.44
2026.04
71.8-
2026.04
71.7971.44
2026.04
71.7470.94
2026.04
71.7471.44
2026.04
71.7465.42
2026.04
71.7-
2026.04
71.5-
2026.04
71.4470.43
2026.04
71.4-
2026.04
71.3870.43
2026.04
71.2470.43
2026.04
71.2-
2026.04
71.1570.94
2026.04
71.1-
2026.04
71.1-
2026.04
71.1-
2026.04
71.1-
2026.04
71-
2026.04
71-
2026.04
70.9-
2026.04
70.670.43
2026.04
70.5570.94
2026.04
69.8170.43
2026.04
68.77-
2026.04
68.32-
2026.04
68.2262.64
2026.04
68.1763.54
2026.04
68.12-
2026.04
68.0164.1
2026.04
67.7262.64
2026.04
67.7263.54
2026.04
67.2863.54
2026.04
67.23-
2026.04
67.0862.64
2026.04
66.78-
2026.04
66.68-
2026.04
66.24-
2026.04
65.8963.54
2026.04
65.4963.54
2026.04
65.49-
2026.04
65.4458.47
2026.04
65.29-
2026.04
65.29-
2026.04
65.29-
2026.04
65.2-
2026.04
65.0562.64
2026.04
64.55-
2026.04
63.86-
2026.04
63.7158.47
2026.04
63.11-
2026.04
62.7755.64
2026.04
62.3762.64
2026.04
61.9258.47
2026.04
60.9365.42
2026.04
60.54-
2026.04
59.49-
2026.04
58.8-
2026.04
58.4-
2026.04
57.6-
2026.04
57.5-
Showing 100 of 131 rows