Update README.md
Browse files
README.md
CHANGED
|
@@ -52,13 +52,14 @@ It is a fine-tune of **Qwen 2.5-VL-7B** using ~10 k synthetic doc-to-Markdown pa
|
|
| 52 |
|
| 53 |
|
| 54 |
### Win-rate of our model against others models:
|
| 55 |
-
|
| 56 |
<img src="bar plot.png" width="500"/>
|
|
|
|
| 57 |
|
| 58 |
### Matrix Win-rate:
|
| 59 |
-
|
| 60 |
<img src="matrix.png" width="500"/>
|
| 61 |
-
|
| 62 |
|
| 63 |
## Training
|
| 64 |
|
|
|
|
| 52 |
|
| 53 |
|
| 54 |
### Win-rate of our model against others models:
|
| 55 |
+
<p align="center">
|
| 56 |
<img src="bar plot.png" width="500"/>
|
| 57 |
+
</p>
|
| 58 |
|
| 59 |
### Matrix Win-rate:
|
| 60 |
+
<p align="center">
|
| 61 |
<img src="matrix.png" width="500"/>
|
| 62 |
+
</p>
|
| 63 |
|
| 64 |
## Training
|
| 65 |
|