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

Multimodal Question Answering on ScienceQA

97.8Accuracy

CASHEW

14.49636.12357.7579.377Dec 5, 2024Feb 23, 2025May 15, 2025Aug 3, 2025Oct 23, 2025Jan 11, 2026Apr 2, 2026
Updated 15d ago

Evaluation Results

MethodLinks
2026.01
97.8--
2026.01
97.8--
2026.01
97.7--
2026.01
96.9--
2026.01
95.9--
2026.01
95.4--
2026.01
93.1--
2026.01
92.9--
2026.01
91.7--
2026.01
88.8--
2025.12
86.28--
2025.12
82.69--
2025.12
81.01--
2026.01
79.1--
2025.12
74.17--
2026.01
73--
2024.12
70.2100-
2026.01
69.5--
2024.12
68.397.3-
2026.01
68.2--
2024.12
68.197-
2024.12
67.996.7-
2024.12
67.796.4-
2024.12
67.596.2-
2024.12
67.596.2-
2024.12
67.596.2-
2024.12
67.395.9-
2024.12
67.395.9-
2026.01
66.8--
2026.04
66.7--
2026.04
65.5--
2025.12
65.45--
2025.12
61.58--
2025.12
61.27--
2025.12
44.87--
2025.12
44.22--
2025.12
38.62--
2026.04
24.6--
2026.04
22.5--
2026.04
21--
2026.04
17.7--
2024.11
--82.4
2024.11
--91.1
2024.11
--92.3
2024.11
--66.6
2024.11
--92.3
2024.11
--90.7
2024.11
--92