Hanging on Robotic Headphone Manipulation Task 1 (test)
1R-LText-guided VLM
Evaluation Results
| Method | Links | ||||
|---|---|---|---|---|---|
| Text-guided VLMChain-of-Thought (COT)=True2025.02 | 1 | — | 81.5 | 38 | |
| Ours2025.02 | 1 | 98.1 | 99 | 96 | |
| Ours (-2D Prompt)2D Prompting=False2025.02 | 1 | 0 | 44.5 | 0 | |
| End2End VLM2025.02 | 0.78 | 0 | 0 | 0 | |
| Text-guided VLM2025.02 | 0.78 | — | 60 | 8.1 | |
| 3D Multimodal LM2025.02 | 0.78 | 52.5 | 64.5 | 13.4 | |
| Ours (-SLM)Supervised Learning Module (SLM)=False2025.02 | 0.78 | 72.3 | 80 | 33.4 | |
| Text-guided LLM2025.02 | 0.683 | — | 43.5 | 0 |