| Task Name | Dataset Name | SOTA Result | Trend | |
|---|---|---|---|---|
| Mobile UI Reasoning | AMEX | Gmail Accuracy76.2 | 17 | |
| GUI Agent Planning and Execution | AMEX (test) | Success Rate (Gmail)77.3 | 12 | |
| GUI Task Execution | Amex | Success Rate67.45 | 8 | |
| Mobile GUI Automation | AMEX | Type Success Rate77.57 | 7 | |
| Action Prediction | AMEX (test) | Action Accuracy72.61 | 4 | |
| Grounding | AMEX (test) | Grounding Score79.19 | 4 | |
| Optical Character Recognition | AMEX (test) | OCR Accuracy79.67 | 4 | |
| Image Captioning | AMEX (test) | Caption Score54.53 | 4 | |
| GUI Navigation | AMEX (High) | Action Matching Score (AMS)80.7 | 3 |