Update app.py

app.py

@@ -204,7 +204,6 @@ def generate_viz_wrapper_t2i(prompt_text, steps, guidance_scale, mask_schedule="
 def generate_viz_wrapper_lm(prompt_text, steps, gen_length, block_length, temperature,
                          cfg_scale, remasking_strategy, thinking_mode_lm=False): 
     global MODEL, TOKENIZER, MASK_ID, DEVICE
-    print(f"thinking_mode_lm: {thinking_mode_lm}")
     if MODEL is None or TOKENIZER is None or MASK_ID is None:
         yield [("Error: Model not loaded. Please load the model first.", "ERROR")], "Model not loaded."
         return
@@ -575,7 +574,6 @@ footer{display:none !important}
 
 # thinking_mode_t2i = gr.State(False)
 def toggle_thinking_mode_lm(current_thinking_mode):
-    # print(f"current_thinking_mode: {current_thinking_mode}")
     new_state = not current_thinking_mode
     new_label = "Thinking Mode ✅" if new_state else "Thinking Mode ❌"
     return new_state, gr.update(value=new_label)    
@@ -599,10 +597,23 @@ with gr.Blocks(css=css_styles, theme=theme) as demo:
 # with gr.Blocks() as demo:
     thinking_mode_lm = gr.State(False)
     thinking_mode_mmu = gr.State(False)
-    gr.Markdown("<h1 style='text-align: center; margin-bottom: 20px;'>MMaDA: Multimodal Large Diffusion Language Models</h1>")
-    gr.Markdown("MMaDA is a novel class of multimodal diffusion foundation models designed to achieve superior performance across diverse domains such as textual reasoning, multimodal understanding, and text-to-image generation")
-    gr.Markdown("Github: [Gen-Verse/MMaDA](https://github.com/Gen-Verse/MMaDA)")
-    gr.Markdown("Paper: [MMaDA: Multimodal Large Diffusion Language Models]()")
+    # gr.Markdown("<h1 style='text-align: center; margin-bottom: 20px;'>MMaDA: Multimodal Large Diffusion Language Models</h1>")
+    # gr.Markdown("MMaDA is a novel class of multimodal diffusion foundation models designed to achieve superior performance across diverse domains such as textual reasoning, multimodal understanding, and text-to-image generation")
+    # gr.Markdown("Github: [Gen-Verse/MMaDA](https://github.com/Gen-Verse/MMaDA)")
+    # gr.Markdown("Paper: [MMaDA: Multimodal Large Diffusion Language Models]()")
+    gr.HTML("""
+    <div align="center" style="margin-bottom: 20px;">
+        <img src="figs/title.png" width="180"/>
+        <p style="font-size: 16px; max-width: 600px; margin: 5px auto;">
+            MMaDA is a new class of multimodal diffusion foundation models, enabling state-of-the-art performance in reasoning, multimodal understanding, and text-to-image generation.
+        </p>
+        <p style="font-size: 15px;">
+            📄 <a href="https://arxiv.org/abs/2505.15809" target="_blank">Paper</a>
+            &nbsp;&nbsp;&nbsp;|&nbsp;&nbsp;&nbsp;
+            💻 <a href="https://github.com/Gen-Verse/MMaDA" target="_blank">Code</a>
+        </p>
+    </div>
+    """)
     with gr.Row():
         model_select_radio = gr.Radio(
             label="Select Text Generation Model",