OSExpert
Benchmarks
| Task Name | Dataset Name | SOTA Result | Trend | |
|---|---|---|---|---|
| Fine-Grained Action Execution | OSExpert-Eval | GIMP Execution Time (s)22 | 10 | |
| Unseen UI Generalization | OSExpert-Eval | Execution Time (Tableau, s)16 | 10 | |
| Fine-Grained Action Execution | OSExpert-Eval | GIMP Success Rate28 | 8 | |
| Unseen UI Generalization | OSExpert Eval | Tableau Success Rate25 | 8 | |
| Long-Horizon Composite Skills | OSExpert-Eval | GIMP Success Rate33 | 8 |
Showing 5 of 5 rows