Mobile Agent Action Execution on OdysseyAgent (random split)
74.98Average Mission Success (AMS)AEIA
Evaluation Results
| Method | Links | ||||||
|---|---|---|---|---|---|---|---|
| AEIA2026.03 | 74.98 | 10.5 | 2.54 | 2.09 | 0.06 | 5.47 | |
| pop-ups2026.03 | 74.98 | 10.5 | 2.48 | 1.78 | 0.1 | 5.57 | |
| Clean2026.03 | 74.88 | 10.2 | 0 | 0 | 0 | 0 | |
| Scenario-BadNets2026.03 | 74.82 | 9.62 | 76.29 | 81.28 | 77.29 | 70.28 | |
| AgentRAE2026.03 | 71.25 | 9.33 | 90.19 | 96.22 | 76.52 | 97.83 |