GUI Interaction on ScreenSpot V1
92Avg AccGUI-G²
Evaluation results
| Method | Links | |||||||
|---|---|---|---|---|---|---|---|---|
| GUI-G²Evaluation Protocol=Upper Bound2026.01 | 92 | 96.7 | 90.8 | 95.9 | 88.6 | 90.9 | 86.9 | |
| SE-GUI†Evaluation Protocol=Upper Bound2026.01 | 88.2 | — | — | — | — | — | — | |
| GUI-AiFContinual Stage=M.->D.->W., Comparison Group=vs GUI-G²2026.01 | 81.7 | 96.1 | 76.9 | 95.7 | 68.3 | 84.8 | 68.2 | |
| GUI-AiFContinual Stage=M.->D., Comparison Group=vs GUI-G²2026.01 | 80.3 | 94.1 | 77.3 | 93.2 | 69.3 | 81.9 | 66.2 | |
| GUI-AiFContinual Stage=M.->D.->W., Comparison Group=vs SE-GUI†2026.01 | 78.9 | 91.9 | 79.5 | 87.1 | 70.7 | 78.2 | 65.8 | |
| GUI-AiFContinual Stage=M., Comparison Group=vs GUI-G²2026.01 | 78.1 | 92.9 | 78.7 | 81.9 | 68.6 | 81.6 | 65.1 | |
| GUI-AiFContinual Stage=M.->D., Comparison Group=vs SE-GUI†2026.01 | 77.4 | 92.7 | 78.2 | 86.6 | 68.4 | 76.2 | 62.2 | |
| GUI-G²Continual Stage=M.->D.->W.2026.01 | 77.1 | 95.2 | 76.4 | 92.8 | 63.6 | 79.6 | 54.9 | |
| GUI-ActorContinual Stage=M.->D.2026.01 | 76.9 | 91.2 | 71.2 | 85.1 | 70 | 74.4 | 69.4 | |
| GUI-AiFContinual Stage=M., Comparison Group=vs SE-GUI†2026.01 | 76.6 | 91.4 | 77.3 | 83.5 | 67.1 | 76.9 | 63.5 | |
| SE-GUI†Continual Stage=M.->D.->W.2026.01 | 76.1 | 92 | 78.2 | 83.1 | 65.7 | 77.4 | 60.2 | |
| GUI-ActorContinual Stage=M.->D.->W.2026.01 | 75 | 87.6 | 70.7 | 87.1 | 67.1 | 72.6 | 65.1 | |
| GUI-G²Continual Stage=M.->D.2026.01 | 74.8 | 93.8 | 76.4 | 88.2 | 61.4 | 75.2 | 53.9 | |
| SE-GUI†Continual Stage=M.->D.2026.01 | 74.4 | 91.8 | 76.6 | 78.1 | 65 | 76.3 | 58.3 | |
| GUI-G²Continual Stage=M.2026.01 | 72.2 | 91.9 | 77.2 | 78.8 | 61.4 | 72.6 | 51.2 | |
| Qwen2.5VL-3BModel Type=Open-source2026.01 | 72.1 | 90.2 | 72.9 | 78.4 | 57.1 | 80.9 | 64.1 | |
| SE-GUI†Continual Stage=M.2026.01 | 71.4 | 91.5 | 75.9 | 72.9 | 59.3 | 73.9 | 55.1 | |
| SeeClickContinual Stage=M.->D.->W.2026.01 | 66.4 | 73.9 | 65.5 | 78.4 | 57.9 | 69.1 | 53.4 | |
| SeeClickContinual Stage=M.->D.2026.01 | 62.1 | 69.1 | 63.4 | 79.8 | 48.5 | 67.2 | 44.3 | |
| GUI-ActorContinual Stage=M.2026.01 | 61 | 74.4 | 76.4 | 58.8 | 61.4 | 32.2 | 63.1 | |
| SeeClickEvaluation Protocol=Upper Bound2026.01 | 53.4 | 78 | 52 | 72.2 | 30 | 55.7 | 32.5 | |
| SeeClickContinual Stage=M.2026.01 | 42.3 | 52.7 | 59.8 | 52.1 | 42.9 | 20.4 | 25.7 | |
| GPT-4oModel Type=Proprietary2026.01 | 18.8 | — | — | — | — | — | — |