GUI Agent Task Completion on WeChat Mini-App Real-World (test)
42.28Success Rate (Human)code-native-reward (Ours)
Evaluation results
| Method | Links | |
|---|---|---|
| code-native-reward (Ours)Training Environment=Synthetic Environment, Hours per step=0.1013, Cost per step=$02026.02 | 42.28 | |
| VLM-reward (Ours)Training Environment=Synthetic Environment, Hours per step=0.1036, Cost per step=$0.52026.02 | 41.61 | |
| VLM-rewardTraining Environment=Real-World Environment, Hours per step=0.2272, Cost per step=$2402026.02 | 40.94 | |
| Base ModelTraining Environment=N/A2026.02 | 36.91 |