Multimodal Text Extraction on LOC
93.51PrecisionSeeAct-Vision
Evaluation results
| Method | Links | |||
|---|---|---|---|---|
| SeeAct-Visionsetting=end-to-end setting2026.05 | 93.51 | 71.63 | 69.46 | |
| Browser-usesetting=end-to-end setting2026.05 | 79.22 | 62.95 | 54.03 | |
| BaRAsetting=end-to-end setting2026.05 | 77.87 | 88.35 | 70.62 | |
| Pure LLMsetting=end-to-end setting2026.05 | 66.11 | 23.71 | 20.54 |