Visual Question Answering on ScienceQA Image (Score)
94.8ScoreHoVLE
Evaluation Results
| Method | Links | |
|---|---|---|
| HoVLE#Params=2.6B, Resolution=Standard2024.12 | 94.8 | |
| HoVLE#Params=2.6B, Resolution=HD2024.12 | 94.8 | |
| HIVEBackbone=Huginn, Param.=4B, Train Data=6.5M, Recurrence Level (r)=322026.02 | 91.6 | |
| Phi3-Vision#Params=4.2B2024.12 | 90.8 | |
| Emu3#Params=8B2024.12 | 89.2 | |
| Emu3Param.=8B2026.02 | 89.2 | |
| InternVL1.5#Params=2.2B2024.12 | 84.9 | |
| InternVL2#Params=2.2B2024.12 | 84.9 | |
| Bunny-v1.1-4BBackbone=Phi-3-3.8B, Param.=4B, Train Data=2.7M2026.02 | 78.3 | |
| Imp-4BBackbone=Phi-3-3.8B, Param.=4B, Train Data=1.6M2026.02 | 78.3 | |
| MobileVLM V2 7BBackbone=Vicuna-7B-v1.5, Param.=7B, Train Data=3.6M2026.02 | 74.8 | |
| SOLO#Params=7B2024.12 | 73.3 | |
| MM1-3B-Chat#Params=3B2024.12 | 69.4 | |
| VILA1.5-3B#Params=3.1B2024.12 | 69 | |
| EVE-7B#Params=7B, Resolution=HD2024.12 | 64.9 | |
| EVE-7B#Params=7B, Resolution=Standard2024.12 | 63 | |
| Chameleon-7B#Params=7B2024.12 | 46 |