| Dataset Name | SOTA Method | Metric | Trend | ||
|---|---|---|---|---|---|
| ActPred | TrimTokenator-LC | Accuracy0.56 | 36 | 4mo ago | |
| DriveAction | Accuracy82.2 | 30 | 18d ago | ||
| AndroidControl Low v2 | GUI-Libra-8B | Pass@1 Step Accuracy88.9 | 22 | 1mo ago | |
| AndroidControl High v2 | Mem-W-8B | Pass@1 Step Accuracy68.59 | 22 | 2mo ago | |
| AITZ (test) | OS-Atlas | Scroll Accuracy76.12 | 17 | 4mo ago | |
| AITZ | ICLAttack | Click TMR95.9 | 15 | 1mo ago | |
| AndroidControl High-level | Click TMR87.4 | 15 | 1mo ago | ||
| AndroidControl Low-level | ICLAttack | TMR (Click)97.6 | 15 | 1mo ago | |
| G3D (test) | SSNet-GT | Accuracy84 | 15 | 4mo ago | |
| Mind2Web Cross-Domain | OmniParser | Operation F185.7 | 14 | 4mo ago | |
| Mind2Web Cross-Task | OmniParser | Operation F1 Score87.6 | 14 | 4mo ago | |
| MiCU Dataset | MiCU-4B-fast | Accuracy (Light)87.21 | 11 | 1mo ago | |
| MIND2WEB Cross-Website 1.0 | HTML-T5-XL | Element Accuracy47.6 | 11 | 4mo ago | |
| MIND2WEB Cross-Task 1.0 | HTML-T5-XL | Element Accuracy60.6 | 11 | 4mo ago | |
| PKUMMD (test) | SSNet-GT | Accuracy (10% Obs)34.8 | 11 | 4mo ago | |
| OAD (test) | SSNet-GT | Accuracy (10% Obs)66.7 | 11 | 4mo ago | |
| ProcTHOR single-object (OOD Systematic) | CDE (ViT-DINO) | Accuracy75 | 9 | 4mo ago | |
| ProcTHOR single-object (OOD Compositional) | CDE (ViT-MAE) | Accuracy95 | 9 | 4mo ago | |
| ProcTHOR single-object (IID) | CDE (ViT-CLIP) | Accuracy97 | 9 | 4mo ago | |
| ScreenSpot Overall | Skill-CMIB | Avg Action Success Rate57.9 | 8 | 2mo ago | |
| ScreenSpot (Web) | Success Rate (Text Actions)70.4 | 8 | 2mo ago | ||
| ScreenSpot Mobile | SeeClick | Action Success Rate (Text)78 | 8 | 2mo ago | |
| GoStanford (test) | RSBM | Action MSE2.95 | 8 | 3mo ago | |
| SCAND (test) | RSBM | Action MSE0.47 | 8 | 3mo ago | |
| SACSoN (test) | RSBM | Action MSE1.32 | 8 | 3mo ago |