3D Object Classification on ModelNet40 60 (test)
70.5Top-1 AccuracyMPM
Evaluation results
| Method | Links | |
|---|---|---|
| MPMTuning=PEFT(78M), Input=3D point cloud2026.05 | 70.5 | |
| PointLLM-7BTuning=Full(~7B), Input=3D point cloud2026.05 | 53.4 | |
| AdapterTuning=PEFT(9.5M), Input=3D point cloud2026.05 | 53.2 | |
| ShapeLLM-7BTuning=Full(~7B), Input=3D point cloud2026.05 | 53.1 | |
| PointLLM-13BTuning=Full(~13B), Input=3D point cloud2026.05 | 53 | |
| ShapeLLM-13BTuning=Full(~13B), Input=3D point cloud2026.05 | 53 | |
| PointBind&LLMTuning=LoRA(~7B), Input=3D point cloud2026.05 | 51.9 | |
| Qwen3-Omni-30B-A3BTuning=–, Input=3-view 2D image2026.05 | 24.2 |