GUI Automation on TreeCUA OOD benchmark 1.0 (test)
3,080SRTreeCUA-DPO-7B
Evaluation results
| Method | Links | |
|---|---|---|
| TreeCUA-DPO-7BBackbone=Qwen2.5-VL-7B, Parameters=7B, Training=Direct Preference Optimization (DPO)2026.02 | 3,080 | |
| TreeCUA-7BBackbone=Qwen2.5-VL-7B, Parameters=7B2026.02 | 2,670 | |
| Qwen2.5-VL-7BModel Family=Qwen, Parameters=7B2026.02 | 80 |