Multi-task policy evaluation on Real World Seen Tasks
46Success Rateπ0.5-DROID-FT
Evaluation Results
| Method | Links | |
|---|---|---|
| π0.5-DROID-FTBase Model=π0.5, Pre-training=DROID, Fine-tuning=SimFoundry2026.06 | 46 | |
| π0.5-FTBase Model=π0.5, Fine-tuning=SimFoundry2026.06 | 45 | |
| π0.5-DROIDBase Model=π0.5, Pre-training=DROID2026.06 | 28 |