| Task Name | Dataset Name | SOTA Result | Trend | |
|---|---|---|---|---|
| Document Visual Question Answering | DocVQA (test) | ANLS96.5 | 192 | |
| Document Visual Question Answering | DocVQA | ANLS95.7 | 164 | |
| Visual Question Answering | DocVQA | Accuracy94.9 | 103 | |
| Document Visual Question Answering | DocVQA | Accuracy94.3 | 81 | |
| Document-Oriented Visual Question Answering | DocVQA | Accuracy94.9 | 72 | |
| Document Visual Question Answering | DocVQA (val) | Accuracy97.85 | 66 | |
| Document Question Answering | DocVQA (test) | Accuracy96.4 | 59 | |
| Document Question Answering | DocVQA | ANLS97.87 | 52 | |
| Document Visual Question Answering | DocVQA v1.0 (test) | ANLS96.5 | 49 | |
| Document Understanding | DocVQA (test) | Accuracy96.5 | 39 | |
| Visual Question Answering | DocVQA | ANLS93.78 | 32 | |
| Visual Question Answering | DocVQA (val) | ANLS89.2 | 31 | |
| Document Visual Question Answering | DocVQA 104 (test) | Score96.1 | 23 | |
| OCR-based Visual Question Answering | DocVQA 2021 (val) | Accuracy93.7 | 13 | |
| Document Understanding | DocVQA | ANLS91.9 | 10 | |
| OCR-related understanding | DocVQA | Score95.1 | 10 | |
| Context Understanding | DocVQA | Accuracy0.994 | 8 | |
| Document and chart understanding | DocVQA | Pass@196.9 | 7 | |
| Question Answering | PFL-DocVQA (test) | ROUGE-10.702 | 7 | |
| Retrieval | MP-DocVQA | R@383.59 | 6 | |
| Visual Question Answering | DocVQA (test) | Accuracy95.9 | 6 | |
| Document Understanding | DocVQA (val) | Accuracy95.45 | 5 | |
| Visual Question Answering | DocVQA (held-out) | ANLS82.8 | 4 | |
| OCR and Chart Understanding | DocVQA | Accuracy95.3 | 3 | |
| Document/Chart Understanding | DocVQA | Score93.6 | 3 |