Update app.py

app.py

@@ -98,7 +98,7 @@ iface = gr.Interface(
     fn=process_smiles,
     inputs=[
         gr.Textbox(label="Input SMILES", value="O1[C@H](CO)[C@@H](O)[C@H](O)[C@@H](O)[C@H]1O[C@@]2(O[C@@H]([C@@H](O)[C@@H]2O)CO)CO"),
-        gr.Checkbox(label="Compute Explanation (Takes 60s)", value=False),
+        gr.Checkbox(label="Display Explanation", value=False),
     ],
     outputs=[
         gr.Label(num_top_classes=3, label="Classification Probabilities"),
@@ -106,7 +106,22 @@ iface = gr.Interface(
         gr.Textbox(label="Canonical SMILES")
     ],
     title="FART",
-    description="Enter a SMILES string to get the taste classification probabilities."
+    description="""
+<section id="molecular-taste-description">
+  <h2>Discover Molecular Taste with FART</h2>
+  <p>
+    At Kvant Labs, we revolutionize taste chemistry with FART (Flavor Analysis and Recognition Transformer), an AI-powered tool designed to predict molecular taste profiles from SMILES. FART delivers predictions for <strong>sweet</strong>, <strong>bitter</strong>, <strong>sour</strong>, and <strong>umami</strong> with over 91% accuracy.
+  </p>
+  <p>
+    Beyond predictions, FART offers interpretable insights by identifying the molecular features driving taste, enabling actionable decisions for flavor innovation. Powered by the ChemBERTa framework and trained on the largest molecular taste dataset to date, FART sets a new standard for precision and efficiency in food science.
+  </p>
+  <p>
+    Learn more about the science behind FART in our <a href="https://chemrxiv.org/engage/chemrxiv/article-details/673a2a3af9980725cf80503c" target="_blank">Pre-print</a>.
+  </p>
+</section>
+""",
+    article=
+    
 )
 
 iface.launch()