Robotic Manipulation on Meta-World
94.7Average Success RateGEA
Evaluation results
| Method | Links | |
|---|---|---|
| GEAProtocol=Zero-shot, Training Scope=Domain Generalist, Input Modalities=Visual2024.12 | 94.7 | |
| [69]Protocol=Zero-shot, Training Scope=Domain Generalist, Input Modalities=Visual2024.12 | 87 | |
| SlotMIMPre-training Dataset=Ego4D, Pre-training Scale=1.28M, Backbone=ViT-B/162025.03 | 84.2 | |
| MLLM+ILProtocol=Zero-shot, Training Scope=Domain Specialist, Input Modalities=Visual2024.12 | 84 | |
| BAKUAction head=MLP2024.06 | 79 | |
| BAKUAction head=VQ-BeT2024.06 | 78 | |
| SlotMIMPre-training Dataset=DetSoup, Pre-training Scale=4M, Backbone=ViT-B/162025.03 | 75.1 | |
| VC-1Pre-training Dataset=Ego4D+MNI, Pre-training Scale=5.6M, Backbone=ViT-B/162025.03 | 73.3 | |
| MVPPre-training Dataset=EgoSoup, Pre-training Scale=4.6M, Backbone=ViT-B/162025.03 | 70.1 | |
| RT-12024.06 | 65 | |
| DINOv2Pre-training Dataset=LVD, Pre-training Scale=142M, Backbone=ViT-B/142025.03 | 38.8 | |
| OATdecoded_tokens=82026.02 | 24.4 | |
| OATdecoded_tokens=42026.02 | 19.5 | |
| Diffusion Policy (DP)2026.02 | 19.3 | |
| QueST2026.02 | 17.9 | |
| OATdecoded_tokens=22026.02 | 16.4 | |
| Bin2026.02 | 14.5 | |
| MT-ACT2024.06 | 13 | |
| OATdecoded_tokens=12026.02 | 11.3 | |
| FAST2026.02 | 7.1 | |
| R3M2024.05 | 0.96 | |
| SCR-FTfine-tuned=true2024.05 | 0.949 | |
| SCRfine-tuned=false2024.05 | 0.944 | |
| VC-12024.05 | 0.923 | |
| CLIP2024.05 | 0.901 | |
| SD-VAE2024.05 | 0.755 | |
| Voltron2024.05 | 0.725 |