| Dataset Name | SOTA Method | Metric | Trend | ||
|---|---|---|---|---|---|
| AndroidWorld | AutoRPA | Overall Success Rate76.9 | 68 | 13d ago | |
| GUI-Odyssey | UI-TARS-7B* | Success Rate (SR)87 | 62 | 1mo ago | |
| AndroidLab | UI-Venus-1.5-8B | Success Rate55.1 | 25 | 1mo ago | |
| PrivScreen | Original | Accuracy93 | 18 | 3mo ago | |
| AMEX | GUI-R1-7B + CES | Type Success Rate77.57 | 7 | 3mo ago | |
| AITZ | GUI-R1-7B + CES | Type Success Rate64.44 | 7 | 3mo ago | |
| AW-Extend | AgentProg | Success Rate68.4 | 6 | 3mo ago |
