update layout

app.py

@@ -131,13 +131,39 @@ def search(query_image, searcher=searcher):
 
 blocks = gr.Blocks()
 
+tldr = """
+We propose two architectures of interpretable image classifiers
+that first explain, and then predict by harnessing 
+the visual correspondences between a query image and exemplars.
+Our models improve on several out-of-distribution (OOD) ImageNet
+datasets while achieving competitive performance on ImageNet
+than the black-box baselines (e.g. ImageNet-pretrained ResNet-50).
+On a large-scale human study (∼60 users per method per dataset)
+on ImageNet and CUB, our correspondence-based explanations led 
+to human-alone image classification accuracy and human-AI team
+accuracy that are consistently better than that of kNN. 
+We show that it is possible to achieve complementary human-AI
+team accuracy (i.e., that is higher than either AI-alone or
+human-alone), on ImageNet and CUB.
+
+<div align="center"> 
+<a href="https://github.com/anguyen8/visual-correspondence-XAI">Github Page</a>
+</div> 
+"""
+
 with blocks:
     gr.Markdown(""" # CHM-Corr DEMO""")
-    gr.Markdown(
-        """ ### Parameters: N=50, k=20 - Using ``ImageNet Pretrained ResNet50`` features"""
-    )
+    gr.Markdown(f""" ## Description: \n {tldr}""")
+
+    with gr.Row():
+        input_image = gr.Image(type="filepath")
+
+        with gr.Column():
+            gr.Markdown(f"### Parameters:")
+            gr.Markdown(
+                "`N=50`\n `k=20` \nUsing `ImageNet Pretrained ResNet50` features"
+            )
 
-    input_image = gr.Image(type="filepath")
     run_btn = gr.Button("Classify")
     gr.Markdown(""" ### CHM-Corr Output Visualization """)
     viz_plot = gr.Image(type="pil", label="Visualization")