Update README.md
Browse files
README.md
CHANGED
|
@@ -9,6 +9,17 @@ LayoutLMv3 is a pre-trained multimodal Transformer for Document AI with unified
|
|
| 9 |
[LayoutLMv3: Pre-training for Document AI with Unified Text and Image Masking](https://arxiv.org/abs/2204.08387)
|
| 10 |
Yupan Huang, Tengchao Lv, Lei Cui, Yutong Lu, Furu Wei, Preprint 2022.
|
| 11 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 12 |
## Citation
|
| 13 |
|
| 14 |
If you find LayoutLM useful in your research, please cite the following paper:
|
| 9 |
[LayoutLMv3: Pre-training for Document AI with Unified Text and Image Masking](https://arxiv.org/abs/2204.08387)
|
| 10 |
Yupan Huang, Tengchao Lv, Lei Cui, Yutong Lu, Furu Wei, Preprint 2022.
|
| 11 |
|
| 12 |
+
## Results
|
| 13 |
+
#### Results on [XFUND](https://github.com/doc-analysis/XFUND)
|
| 14 |
+
| Language | Precision | Recall | F1 |
|
| 15 |
+
|-----------|------------|------|--------|
|
| 16 |
+
| ZH | 0.8910 | 0.9374 | 0.9136 |
|
| 17 |
+
|
| 18 |
+
#### Results on [EPHOIE](https://github.com/HCIILAB/EPHOIE)
|
| 19 |
+
| Subject | Test Time | Name | School | Examination Number | Seat Number | Class | Student Number | Grade | Score | **Mean** |
|
| 20 |
+
|:------------|:------:|:--------:|:--------:|:--------:|:--------:|:--------:|:--------:|:--------:|:--------:|:--------:|
|
| 21 |
+
| 98.48 | 100 | 99.36 | 98.86 | 100 | 100 | 98.73 | 98.89 | 97.59 | 97.78 | 98.97 |
|
| 22 |
+
|
| 23 |
## Citation
|
| 24 |
|
| 25 |
If you find LayoutLM useful in your research, please cite the following paper:
|