Widen title rules, remove HF link, add demo screenshot

README.md

@@ -13,10 +13,10 @@ library_name: transformers
 <div align="center">
   <h3>
     <b>
-      <span>━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━</span><br/>
+      <span>━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━</span><br/>
       MiMo-V2.5-ASR: Robust Speech Recognition Across<br/>
       Languages, Dialects, and Complex Acoustic Scenarios<br/>
-      <span>━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━</span>
+      <span>━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━</span>
     </b>
   </h3>
 </div>
@@ -25,8 +25,6 @@ library_name: transformers
 
 <div align="center" style="line-height: 1;">
   |
-  <a href="https://huggingface.co/collections/XiaomiMiMo/mimo-v2.5-asr" target="_blank">🤗 HuggingFace</a>
-  &nbsp;|
   <a href="https://github.com/XiaomiMiMo/MiMo-V2.5-ASR" target="_blank">💻 GitHub</a>
   &nbsp;|
   <a href="https://xiaomimimo.github.io/MiMo-V2.5-ASR-Demo" target="_blank">📰 Blog</a>
@@ -126,6 +124,8 @@ This launches a local Gradio interface for MiMo-V2.5-ASR. You can:
 * Optionally specify a **language tag** (Chinese / English / Auto) to bias the model for a specific language, or leave it to **Auto** for automatic language detection (recommended for code-switched speech).
 * The demo calls the `asr_sft()` interface under the hood.
 
+![MiMo-V2.5-ASR Gradio Demo](https://huggingface.co/XiaomiMiMo/MiMo-V2.5-ASR/resolve/main/assets/MiMo_ASR_Demo.png)
+
 To load the model and tokenizer automatically at startup, pass their paths on the command line:
 
 ```bash