3D Visual Question Answering on Hypo3D
35.9EM@1POMA-3D_llm
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| POMA-3D_llmModality=PM, Grid inputs=H x W x 3, LoRA-tuned=true2025.11 | 35.9 | — | |
| UniScene3DInput=CPM2026.04 | 35.2 | 85.9 | |
| POMA-3DInput=PM2026.04 | 33.4 | 84.8 | |
| POMA-3D_specModality=PM2025.11 | 33.4 | 84.8 | |
| LLaVA-OV-7BModality=RGB2025.11 | 33.2 | — | |
| LLaVA-3DModality=RGB-D2025.11 | 33.1 | — | |
| SceneVerseInput=PC2026.04 | 31.6 | 80.3 | |
| SceneVerseModality=C-PC2025.11 | 31.6 | 80.3 | |
| FG-CLIPInput=I2026.04 | 31.1 | 82.1 | |
| FG-CLIPModality=PM2025.11 | 31.1 | 82.1 | |
| 3D-VisTAInput=PC2026.04 | 31 | 81.2 | |
| 3D-ViSTAModality=C-PC2025.11 | 31 | 81.2 | |
| Qwen2.5-VL-7BModality=RGB2025.11 | 30.9 | — | |
| LEOModality=C-PC2025.11 | 16.2 | — |