Visual Question Answering on TextVQA (EM, F1)
32.76EMPixSearch-13B
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| PixSearch-13BBackbone=LLaVA-13B, Search Strategy=No Search2026.01 | 32.76 | 34.55 | |
| PixSearch-13BBackbone=LLaVA-13B, Search Strategy=Full2026.01 | 32.44 | 35.1 | |
| PixSearch-13BBackbone=LLaVA-13B, Search Strategy=Whole Image2026.01 | 32.3 | 34.13 | |
| PixSearch-13BBackbone=LLaVA-13B, Search Strategy=Question2026.01 | 30.87 | 32.9 | |
| PLUM-13BBackbone=LLaVA-13B2026.01 | 30.11 | 32.62 | |
| LLaVA-13BBackbone=LLaVA-13B2026.01 | 22.84 | 26.15 |