UI/UX selection on WiserUI-Bench
62.1FAPERCEPTUI
Evaluation results
| Method | Links | ||||
|---|---|---|---|---|---|
| PERCEPTUISetting=-2026.06 | 62.1 | 86.4 | 74.25 | 44.3 | |
| PERCEPTUISetting=w/o RPE2026.06 | 59.8 | 85.2 | 72.5 | 41.1 | |
| InternVL-2.5-38BSetting=Zero-shot2026.06 | 55.67 | 60 | 57.83 | 34.56 | |
| LLaVA-NeXT-7BSetting=Zero-shot2026.06 | 54.44 | 40.56 | 47.5 | 10.78 | |
| GPT-5Setting=MAD (R1)2026.06 | 53.78 | 69.62 | 61.7 | 40.56 | |
| RandomSetting=-2026.06 | 50 | 50 | 50 | 25 | |
| Claude Opus 4.6Setting=MAD (R1)2026.06 | 42.33 | 74.07 | 58.2 | 33.11 | |
| GPT-5Setting=DDCoT2026.06 | 38.13 | 87.87 | 63 | 36.2 | |
| PERCEPTUISetting=w/o CR2026.06 | 37.4 | 73.8 | 55.6 | 37.2 | |
| GPT-5Setting=CoCoT2026.06 | 35.22 | 86.14 | 60.68 | 32.78 | |
| GPT-5Setting=Self-Refine2026.06 | 34 | 85.34 | 59.67 | 31.44 | |
| GPT-5Setting=Zero-shot2026.06 | 33.91 | 89.8 | 61.86 | 31.09 | |
| Qwen2.5-VL-32BSetting=Zero-shot2026.06 | 33.67 | 81.44 | 57.56 | 31 | |
| GPT-4oSetting=Zero-shot2026.06 | 31.89 | 88.33 | 60.11 | 30.11 | |
| Claude Opus 4.6Setting=Zero-shot2026.06 | 28.13 | 84.55 | 56.34 | 26.44 | |
| LLaVA-OneVision-7BSetting=Zero-shot2026.06 | 19.78 | 81.56 | 50.67 | 10.44 | |
| o1Setting=Zero-shot2026.06 | 16.56 | 97.78 | 57.17 | 15.56 |