3D Question Answering on ScanQA v1.0 (test)
23BLEU-4C2RoPE
Evaluation results
| Method | Links | ||||||
|---|---|---|---|---|---|---|---|
| C2RoPEModel Category=Ours2026.02 | 23 | 109.8 | 34.1 | 52.6 | 31.3 | — | |
| GR3D-8BParameters=8B, Backbone=Qwen-2-7B, Input Resolution=4482026.05 | 18.1 | 105.1 | — | — | 29.2 | — | |
| Ross3DModel Category=3D LLMs2026.02 | 17.9 | 107 | 20.9 | 50.7 | 30.8 | — | |
| 3D-LLaVA2026.05 | 17.1 | 92.6 | — | — | — | — | |
| Video-3D-LLMModel Category=3D LLMs2026.02 | 16.4 | 102.1 | 20 | 49.3 | 30.1 | — | |
| Video-3D LLM2026.05 | 16.2 | 102.1 | — | — | 30.1 | — | |
| PAVE-7B (w/ 3D info)Mode=Task-specific, TFLOPS=98.68, Total Params=8.2B, Trainable Params=170.3M2025.03 | 16 | 103.4 | 19.9 | 49 | 29.1 | 48.5 | |
| SPAR2026.05 | 15.3 | 90.7 | — | — | 27.7 | — | |
| LLaVA-3D-7BMode=Task-specific2025.03 | 14.5 | 91.7 | 20.7 | 50.1 | 27 | 45 | |
| LLaVA-3DModel Category=3D LLMs2026.02 | 14.5 | 91.7 | 20.7 | 50.1 | 27 | — | |
| ChatSceneModel Category=3D LLMs2026.02 | 14.3 | 87.7 | 18 | 41.6 | 21.6 | — | |
| LLaVA-OV-7B-FTMode=Task-specific, TFLOPS=98.53, Total Params=8.2B, Trainable Params=161.5M2025.03 | 13.5 | 95.1 | 19.1 | 47.4 | 27.4 | 46.3 | |
| Grounded 3D-LLMModel Category=3D LLMs2026.02 | 13.4 | — | — | 72.7 | — | — | |
| AdaToken-3DModel Category=3D LLMs2026.02 | 13.3 | 91.8 | 17.6 | 44.7 | 30 | — | |
| LEO-7BMode=Task-specific2025.03 | 13.2 | 101.4 | 20 | 49.2 | 24.5 | 47.6 | |
| LEOModel Category=3D LLMs2026.02 | 13.2 | 101.4 | 20 | 49.2 | 24.5 | — | |
| 3D-LLM-7BMode=Task-specific2025.03 | 12.9 | 74.5 | 15.1 | 37.5 | 21.2 | — | |
| Scene-LLM-7BMode=Task-specific2025.03 | 12 | 80 | 16.6 | 40 | 27.2 | — | |
| 3D-LLMModel Category=3D LLMs2026.02 | 12 | 69.4 | 14.5 | 35.7 | 20.5 | — | |
| Scene-LLMModel Category=3D LLMs2026.02 | 12 | 80 | 16.6 | 40 | 27.2 | — | |
| 3D-VLPModel Category=Expert Models2026.02 | 11.2 | — | 13.5 | 34.5 | 21.1 | — | |
| ScanQAModel Category=Expert Models2026.02 | 10.1 | 64.9 | 13.1 | 33.3 | 21.1 | — | |
| VideoChat2-7BMode=Zero-shot2025.03 | 9.6 | 49.2 | 9.5 | 28.2 | 19.2 | — | |
| Chat-3DModel Category=3D LLMs2026.02 | 6.4 | 53.2 | 11.9 | 28.5 | — | — | |
| LLaVA-OV-7BMode=Zero-shot, TFLOPS=98.53, Total Params=8.2B2025.03 | 5.3 | 91 | 18.2 | 45.9 | 26.7 | 44.3 | |
| InternVl2-8BModel Category=2D LLMs2026.02 | 3.3 | 62.5 | 14.5 | 34.3 | — | — | |
| LLaVA-Video-7BModel Category=2D LLMs2026.02 | 3.1 | 88.7 | 17.7 | 44.6 | — | — | |
| Qwen2-Vl-7BModel Category=2D LLMs2026.02 | 3 | 53.9 | 11.4 | 29.3 | — | — | |
| 3D-VisTAModel Category=Expert Models2026.02 | — | — | 13.9 | 35.7 | 22.4 | — | |
| LL3DAModel Category=3D LLMs2026.02 | — | 80 | 15.9 | 35.9 | — | — |