Single-round Q&A on ShapeNeRF-Text (test)
81.03S-BERT Similarity ScoreLLaNA-7b
Evaluation results
| Method | Links | |||||
|---|---|---|---|---|---|---|
| LLaNA-7bModality=NeRF2024.06 | 81.03 | 81.56 | 46.16 | 53.17 | 50.15 | |
| LLaNAArchitecture setting=single-architecture, Trained on=MLP2025.02 | 81 | 81.6 | — | — | — | |
| LR+CArchitecture setting=single-architecture, Trained on=MLP2025.02 | 81 | 81.6 | — | — | — | |
| PointLLM-7bModality=Point cloud2024.06 | 74.7 | 74.4 | 36.81 | 44.41 | 39.76 | |
| LLaVA-vicuna-13bModality=Image (multi-view)2024.06 | 71.84 | 71.16 | 20.04 | 30.2 | 33.46 | |
| LLaVA-vicuna-7bModality=Image (front-view)2024.06 | 71.79 | 71.96 | 25.79 | 34.04 | 34.86 | |
| LLaVA-vicuna-13bModality=Image (front-view)2024.06 | 71.61 | 70.98 | 20.19 | 30.42 | 32.53 | |
| LLaVA-vicuna-7bModality=Image (back-view)2024.06 | 70.88 | 70.93 | 25.17 | 33.3 | 34.22 | |
| 3D-LLMModality=Mesh + Multi-view2024.06 | 69.62 | 67.55 | 32.19 | 40.95 | 35.83 | |
| LLaVA-vicuna-13bModality=Image (back-view)2024.06 | 68.25 | 69.06 | 20.03 | 29.84 | 32.27 | |
| BLIP-2 FlanT5-xxlModality=Image (front-view)2024.06 | 45.2 | 47.92 | 11.5 | 20.16 | 13.49 | |
| BLIP-2 FlanT5-xxlModality=Image (back-view)2024.06 | 45.06 | 47.66 | 11.5 | 19.98 | 13.44 | |
| GPT4Point-Opt-2.7bModality=Point cloud2024.06 | 27.62 | 31.41 | 6.26 | 9.38 | 5.41 |