| Dataset Name | SOTA Method | Metric | Trend | ||
|---|---|---|---|---|---|
| OOD Evaluation Suite (ScreenSpot-Pro, OSWorld-G, ScreenSpot-V2) unweighted mean | Click-in-Target Accuracy (SS-Pro)60 | 11 | 1mo ago | ||
| WinDOM in-distribution 2,000-example subsample (held-out) | Click-in-target Accuracy82.1 | 11 | 1mo ago |