Generalist Robotic Policy Evaluation on RoboCasa GR1 Tabletop (simulation)
56.5Average Success RateCoRE-VLA
Evaluation Results
| Method | Links | |
|---|---|---|
| CoRE-VLAevaluation_protocol=mean ± std over 5 groups of 100 rollouts2026.07 | 56.5 | |
| GR00T-N1.52026.07 | 48 | |
| GR00T-N1.62026.07 | 47.6 | |
| Diffusion Policy2026.07 | 40.4 |