modifying readme

README-zh_CN.md

@@ -149,7 +149,7 @@ PDF内容提取框架如下图所示
 <span id="mfd-anchor"></span>
 ### 公式检测
 
-我们与开源的模型[Pix2Text-MFD](https://github.com/breezedeus/pix2text)做了对比，另外，Yolov8-SFT是我们在Yolov8模型的基础上进行了SFT训练后的权重。论文验证集由255张Arxiv论文页面构成，多源验证集由789张不同来源的页面构成，包括教材、书籍等。
+我们与开源的模型[Pix2Text-MFD](https://github.com/breezedeus/pix2text)做了对比，另外，Yolov8-SFT是我们在Yolov8模型的基础上进行了SFT训练后的权重。论文验证集由255张论文页面构成，多源验证集由789张不同来源的页面构成，包括教材、书籍等。
 
 <table>
     <tr>

README.md

@@ -81,7 +81,7 @@ Existing open-source models are often trained on data from Arxiv papers and fall
 <span id="layout-anchor"></span>
 ### Layout Detection
 
-We have compared our model with existing open-source layout detection models, including [DocXchain](https://github.com/AlibabaResearch/AdvancedLiterateMachinery/tree/main/Applications/DocXChain), [Surya](https://github.com/VikParuchuri/surya), and two models from [360LayoutAnalysis](https://github.com/360AILAB-NLP/360LayoutAnalysis). The model present as LayoutLMv3-SFT in the table refers to the checkpoint we further trained with our SFT data on [LayoutLMv3](https://github.com/microsoft/unilm/blob/master/layoutlmv3). The validation set for academic papers consists of 402 pages from Arxiv, while the textbook validation set is composed of 587 pages from various sources of textbooks.
+We have compared our model with existing open-source layout detection models, including [DocXchain](https://github.com/AlibabaResearch/AdvancedLiterateMachinery/tree/main/Applications/DocXChain), [Surya](https://github.com/VikParuchuri/surya), and two models from [360LayoutAnalysis](https://github.com/360AILAB-NLP/360LayoutAnalysis). The model present as LayoutLMv3-SFT in the table refers to the checkpoint we further trained with our SFT data on [LayoutLMv3](https://github.com/microsoft/unilm/blob/master/layoutlmv3). The validation set for academic papers consists of 402 pages, while the textbook validation set is composed of 587 pages from various sources of textbooks.
 
 <table>
     <tr>