Add paper link

app.py

@@ -93,7 +93,7 @@ with gr.Blocks() as demo:
         gr.HTML(
             f'''<h1 text-align="center">Creativity Index</h1>
 
-            <p style='font-size: 16px;'>Compute the <a href="">Creativity Index</a> of a piece of text.</p>
+            <p style='font-size: 16px;'>Compute the <a href="https://arxiv.org/pdf/2410.04265">Creativity Index</a> of a piece of text.</p>
             <p style='font-size: 16px;'>The Creativity Index is computed based on verbatim matching against massive text corpora and is powered by <a href="https://infini-gram.io">infini-gram</a>. It is defined as the ratio of tokens not covered by n-grams (n >= L) that can be found in the corpus, averaged across {NGRAM_LEN_MIN} <= L <= {NGRAM_LEN_MAX}. You can view the covered tokens (highlighted in red background) for each value of L.</p>
             <p style='font-size: 16px;'><b>Note:</b> The input text is limited to {MAX_QUERY_CHARS} characters. Each query has a timeout of {MAX_TIMEOUT_IN_SECONDS} seconds. If you have waited 30 seconds and receive an error, you can try submitted the same query and it's more likely to work on the second try.</p>
             <p style='font-size: 16px;'><b>Disclaimer 1:</b> The Creativity Index of text that appear exactly in the corpora may be deflated. In our paper, we remove exact duplicates (including quotations and citations) from the corpus before computing the Creativity Index. However, deduplication is not applied in this demo.</p>