Update app.py

app.py

@@ -60,9 +60,8 @@ inputs = [input_img, input_txt]
 outputs = [gr.inputs.Image(type='pil', label="Output Image"), "highlight"]
 
 
-description = """A demonstration based on the Generic Attention-model Explainability method for Interpreting Bi-Modal
-                 Transformers by Chefer et al. (2021): https://github.com/hila-chefer/Transformer-MM-Explainability.
-                 <br> <br>
+description = """This demo is a copy of the demo CLIPGroundingExlainability built by Paul Hilders, Danilo de Goede and Piyush Bagad, as part of the course Interpretability and Explainability in AI (MSc AI, UvA, June 2022).
+<br> <br>
                  This demo shows attributions scores on both the image and the text input when presenting CLIP with a
                  <text,image> pair. Attributions are computed as Gradient-weighted Attention Rollout (Chefer et al.,
                  2021), and can be thought of as an estimate of the effective attention CLIP pays to its input when
@@ -155,4 +154,4 @@ with demo_tabs:
                 \[2\]: Abnar, S., & Zuidema, W. (2020). Quantifying attention flow in transformers. arXiv preprint arXiv:2005.00928. <br>
                 \[3\]: [https://samiraabnar.github.io/articles/2020-04/attention_flow](https://samiraabnar.github.io/articles/2020-04/attention_flow) <br>
                 """)
-demo_tabs.launch(show_error=True)
+demo_tabs.launch(show_error=True)