Update model description and enhance demo details in app.py

app.py

@@ -338,6 +338,7 @@ title_html = """
 <div style="text-align: center;">
     <img src="https://cdn-uploads.huggingface.co/production/uploads/6336b5c831efcb5647f00170/-G297bBqMzYvTbD6_Bkd9.png" style="height: 95px; width: 100%;">
     <p style="font-size: 20px;">❄️Vintern-1B-v3_5❄️</p>
+    <p style="font-size: 14px;">An Efficient Multimodal Large Language Model for Vietnamese🇻🇳</p>
     <a href="https://huggingface.co/papers/2408.12480" style="font-size: 13px;">[📖 Vintern Paper]</a>
     <a href="https://huggingface.co/5CD-AI" style="font-size: 13px;">[🤗 Huggingface]</a>
 </div>
@@ -345,7 +346,7 @@ title_html = """
 
 description_html = """
 <div style="text-align: center;">
-    <p style="font-size: 14px;">An Efficient Multimodal Large Language Model for Vietnamese🇻🇳</p>
+    <p style="font-size: 12px;">Vintern-1B-v3.5 is the latest in the Vintern series, bringing major improvements over v2 across all benchmarks. This continuous fine-tuning Version enhances Vietnamese capabilities while retaining strong English performance. It excels in OCR, text recognition, and Vietnam-specific document understanding.</p>
 </div>
 """