Update Visual-CoT demo - 2025-10-12 23:47:44

Fixes:
- Fix LLaVA config registration error (compatibility with newer transformers)
- Update Gradio to latest version (security fixes)
- Auto-deployed via update script

app.py

@@ -685,15 +685,40 @@ def create_demo():
                             visible=False,
                         )
                 
-                # Example images
-                gr.Markdown("### 📋 Try These Examples")
+                # Example questions (20 diverse examples)
+                gr.Markdown("### 📋 Try These Example Questions")
                 gr.Examples(
                     examples=[
+                        # Available images
                         ["examples/extreme_ironing.jpg", "What is unusual about this image?"],
                         ["examples/waterview.jpg", "What are the things I should be cautious about when I visit here?"],
+                        # Visual reasoning examples (upload your own images)
+                        [None, "What color is the car in the image?"],
+                        [None, "How many people are in this picture?"],
+                        [None, "What is the main object in the center of the image?"],
+                        [None, "What is the person doing in this photo?"],
+                        [None, "What time of day does this appear to be?"],
+                        [None, "What is the weather like in this image?"],
+                        [None, "What room is this photo taken in?"],
+                        [None, "What brand or logo can you see?"],
+                        # Text reading examples
+                        [None, "What text is written on the sign?"],
+                        [None, "What is the price shown in the image?"],
+                        [None, "What does the document say?"],
+                        [None, "What is the title of this book/poster?"],
+                        # Spatial reasoning
+                        [None, "What is to the left of the main object?"],
+                        [None, "What is on top of the table?"],
+                        [None, "Where is the person standing?"],
+                        # Scene understanding
+                        [None, "What type of place is this?"],
+                        [None, "What activity is happening here?"],
+                        [None, "What is the overall mood or atmosphere?"],
+                        [None, "What can you infer about the context of this image?"],
                     ],
                     inputs=[image_input, question_input],
-                    label="Click to load example",
+                    label="Click to load example questions (upload image for questions without images)",
+                    examples_per_page=10,
                 )
                 
                 # Event handlers