List your company

Multimodal Science Question Answering on ScienceQA

88.4Overall Average ScoreHuman
49.71259.75669.879.844Dec 3, 2025Dec 11, 2025Dec 20, 2025Dec 29, 2025Jan 7, 2026Jan 16, 2026Jan 25, 2026

Evaluation results

MethodLinks
Human2025.1288.490.2384.9787.4889.687.588.191.5982.42
GPT-4o-mini (MemVerse)2025.1285.4885.2681.5589.0983.2878.1991.588.1180.75
CoT (GPT-4)#T-Param=1T+2025.1283.9985.4872.4490.2782.6571.4992.8986.6679.04
HoT-T5-Large#T-Param=738M2025.1283.3884.4679.0884.6482.8975.8188.1583.8882.47
GPT-42025.1282.6984.0673.4587.3681.8770.7590.7384.6979.1
Qwen2.5-72B (MemVerse)#T-Param=72B2025.1280.2577.5368.9585.3678.6866.3989.282.3177.76
DDCoT#T-Param=175B2025.1280.1580.1576.7282.8278.8972.5385.0282.8675.21
Chameleon (ChatGPT)#T-Param=175B+2025.1279.9381.6270.648479.7770.886.6281.8676.53
Qwen2.5-72B#T-Param=72B2025.1278.3779.6467.184.977.566587.9380.2574.85
LaVIN-13B#T-Param=13B2025.1277.54————————
CCOT#T-Param=7B2025.1276.84————————
GPT-4o-mini2025.1276.8277.3173.4586.9174.0566.8687.9383.3771.85
Qwen2.5-7B (MemVerse)#T-Param=7B2025.1275.6274.5168.578.7375.9266.1981.9579.764.73
CoT (GPT-3)#T-Param=173B2025.1275.1775.4470.8778.0974.6867.4379.9378.2369.68
Qwen2.5-7B (Visual Genome)#T-Param=7B2025.1275.0876.7867.0478.0974.0566.1979.7278.0869.68
Qwen2.5-7B#T-Param=7B2025.1274.7276.267.8377.2774.4965.7979.0277.7269.35
BLIP-22025.1274.17————————
CoT (UnifiedQA)#T-Param=223M2025.1274.117176.0478.9166.4266.5381.8177.0668.82
Qwen2.5-7B (Mmkg)#T-Param=7B2025.1273.4773.9866.3778.1871.6564.379.6576.5168.03
GraphVis#T-Param=7B2025.1273.18————————
Vanilla (LLaVA-1.5-7B)Retained Tokens=5762026.0169.5————————
PyramidDropRetained Tokens=1922026.0169.2————————
SparseVLMRetained Tokens=1922026.0169.2————————
VisionZipRetained Tokens=642026.0169————————
VisionZipRetained Tokens=1922026.0168.9————————
VisionZipRetained Tokens=1282026.0168.9————————
ViTCoPRetained Tokens=642026.0168.8————————
ViTCoPRetained Tokens=1922026.0168.4————————
PyramidDropRetained Tokens=1282026.0168.4————————
ViTCoPRetained Tokens=1282026.0168.4————————
FastVRetained Tokens=1922026.0167.2————————
SparseVLMRetained Tokens=1282026.0167.2————————
SparseVLMRetained Tokens=642026.0162.2————————
PyramidDropRetained Tokens=642026.0160.8————————
FastVRetained Tokens=1282026.0160.2————————
FastVRetained Tokens=642026.0151.2————————