Mobile Agent Action Execution on OdysseyAgent task split
56.18Average Mobile Success (AMS)AEIA
Evaluation Results
| Method | Links | ||||||
|---|---|---|---|---|---|---|---|
| AEIA2026.03 | 56.18 | 0 | 1.4 | 2.16 | 0.03 | 2.01 | |
| pop-ups2026.03 | 56.18 | 0 | 1.36 | 2.05 | 0.03 | 2.01 | |
| Clean2026.03 | 56.14 | 0 | 0 | 0 | 0 | 0 | |
| Scenario-BadNets2026.03 | 56.05 | 59 | 36.55 | 28.68 | 36.88 | 44.1 | |
| AgentRAE2026.03 | 56 | 0 | 91.93 | 95.2 | 88.6 | 92 |