Update app.py

app.py

@@ -172,14 +172,14 @@ def inference(img):
   
       sents = utils.decode_sequence(model.vocab, seq)
 
-      return sents
+      return sents[0]
 
 demo = gr.Blocks()
 
 with demo:
     gr.Markdown(
     """
-    # Demo for CLIP-Caption-Reward
+    # Gradio Demo for [j-min/CLIP-Caption-Reward](https://github.com/j-min/CLIP-Caption-Reward)
     """)
     inp = gr.Image(type="pil")
     out = gr.Textbox()