Update app.py

app.py

@@ -284,7 +284,7 @@ By [Princeton Language and Intelligence (PLI), Princeton University](https://pli
 
 ### This is a demonstration of the Skill-Mix evaluation. 
 
-Paper link: [coming soon](www.arxiv.org)
+Paper link: [https://arxiv.org/abs/2310.17567](https://arxiv.org/abs/2310.17567)
 
 ### Samples are generated using 10% of the full set of skills and topics. Click the second tab for comparison between two generations.
 
@@ -427,9 +427,32 @@ Coming soon: generation by more models; grading by LLaMA-2.
                     c.change(fn_list[0], input_list[0], output_list[0]).then(fn_list[1], input_list[1], output_list[1]).then(fn_list[2], input_list[2], output_list[2]).then(fn_list[3], input_list[3], output_list[3]).then(fn_list[4], input_list[4], output_list[4]).then(fn_list[5], input_list[5], output_list[5])
                 else:
                     raise NotImplementedError
-        gr.Markdown('''
+        gr.Markdown('''Please consider citing
 ```
-x
+@article{yu2023skillmix,
+      title={Skill-Mix: a Flexible and Expandable Family of Evaluations for AI models}, 
+      author={Yu, Dingli and Kaur, Simran and Gupta, Arushi and Brown-Cohen, Jonah and Goyal, Anirudh and Arora, Sanjeev},
+      journal={arXiv preprint arXiv:2310.17567},
+      year={2023}
+}
+```
+```
+@misc{openai2023gpt4,
+      title={GPT-4 Technical Report}, 
+      author={OpenAI},
+      year={2023},
+      eprint={2303.08774},
+      archivePrefix={arXiv},
+      primaryClass={cs.CL}
+}
+```
+```
+@article{touvron2023llama,
+  title={Llama 2: Open foundation and fine-tuned chat models},
+  author={Touvron, Hugo and Martin, Louis and Stone, Kevin and Albert, Peter and Almahairi, Amjad and Babaei, Yasmine and Bashlykov, Nikolay and Batra, Soumya and Bhargava, Prajjwal and Bhosale, Shruti and others},
+  journal={arXiv preprint arXiv:2307.09288},
+  year={2023}
+}
 ```
         ''')
     return demo