Robot Control on Real-world Tabletop Manipulation
100T1 ScoreVISCOP
Evaluation Results
| Method | Links | |||||
|---|---|---|---|---|---|---|
| VISCOPTraining data=VIMA-Bench + xArm-Det2025.10 | 100 | 100 | 90 | 96.67 | 96.67 | |
| Adaptation Strategy (Trainable VE, Trainable LLM)Training data=VIMA-Bench + xArm-Det, VE=Trainable, LLM=Trainable2025.10 | 85 | 85 | 70 | 80 | 80 | |
| Adaptation Strategy (Trainable VE, Trainable LLM)Training data=VIMA-Bench, VE=Trainable, LLM=Trainable2025.10 | 45 | 60 | 15 | 40 | 40 | |
| VISCOPTraining data=VIMA-Bench2025.10 | 40 | 70 | 20 | 43.33 | 43.33 |