Failure-Aware Test Generation on OpenVLA-7b in SimplerEnv (test)
95.3Failure Rate (%)Random
Evaluation Results
| Method | Links | ||||
|---|---|---|---|---|---|
| RandomMethod Class=Baseline (B1)2026.06 | 95.3 | 66.8 | 63.2 | 100 | |
| FSCSARTMethod Class=Baseline (B2)2026.06 | 95.7 | 67.8 | 65.4 | 100 | |
| Weighted_RFAlgorithm=Weighted_RF2026.06 | 97.5 | — | — | — | |
| Sorting_RFAlgorithm=Sorting_RF2026.06 | — | 70.5 | — | — | |
| Sorting_RF / Weighted_RFAlgorithm=Sorting_RF / Weighted_RF2026.06 | — | — | 70.1 | — |