Critic Accuracy on RoboCasa OOD Only v1
97Countr Cab AccuracyRobot Critics
Evaluation results
| Method | Links | ||||||||
|---|---|---|---|---|---|---|---|---|---|
| Robot CriticsTraining Data=Full (Success and Failure), Mode=Fine-tuned2026.06 | 97 | 90.9 | 86.3 | 95.8 | 87 | 90.3 | 92.4 | 91.4 | |
| Robot CriticsTraining Data=Success-Only, Mode=Fine-tuned2026.06 | 94 | 85.1 | 82.9 | 89.4 | 79.2 | 89.7 | 89.6 | 87.1 | |
| ProgressLMMode=Fine-tuned2026.06 | 20 | 49.5 | 70.4 | 84.3 | 50.2 | 77 | 68 | 59.9 |