Rollout-level failure prediction on LIBERO-10 (test)
0.838AUROCSW+ATR+BO
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| SW+ATR+BOparameters=w*, alpha*, beta*2026.03 | 0.838 | 74.3 | |
| SW+ATRw=60, alpha=0.92026.03 | 0.738 | 71.1 | |
| Sliding Window Poolingw=702026.03 | 0.699 | 69.1 | |
| Action Transfer Reweightingalpha=0.92026.03 | 0.508 | 60.6 | |
| Baselineparams=mean_{t,d}2026.03 | 0.468 | 62.2 |