update space description and links

app.py

@@ -170,24 +170,24 @@ def gradio_answer(chatbot, chat_state, img_list, num_beams, temperature):
     return chatbot, chat_state, img_list
 
 title = """
-<h1 align="center">Global-Local QFormer for Long Video Understanding with LLMs</h1>
+<h1 align="center">Koala: Key frame-conditioned long video-LLM</h1>
 
-<h5 align="center">  Introduction: We introduce a Global-Local QFormer video model that is connected with a Large Language Model to understand and answer questions about long videos. To try out this demo, please upload a video and start the chat. </h5> 
+<h5 align="center">  Introduction: We introduce a key frame-conditioned video model that is connected with a Large Language Model to understand and answer questions about long videos. To try out this demo, please upload a video and start the chat. </h5> 
 
 <div style='display:flex; gap: 0.25rem; '>
-<a href='https://huggingface.co/spaces/rxtan/Global-Local-QFormer-Video-LLM'><img src='https://img.shields.io/badge/%F0%9F%A4%97%20Hugging%20Face-Spaces-blue'></a> 
-<a href='https://cs-people.bu.edu/rxtan/projects/Global-Local-QFormer/pdf/CVPR_2024_paper.pdf'><img src='https://img.shields.io/badge/Paper-PDF-red'></a>
+<a href='https://huggingface.co/spaces/rxtan/Koala-video-llm'><img src='https://img.shields.io/badge/%F0%9F%A4%97%20Hugging%20Face-Spaces-blue'></a> 
+<a href='https://cs-people.bu.edu/rxtan/projects/Koala/pdf/cvpr_2024_koala.pdf'><img src='https://img.shields.io/badge/Paper-PDF-red'></a>
 </div>
 
 
-Thank you for using the Global-Local QFormer Demo Page! If you have any questions or feedback, please feel free to contact us. 
+Thank you for using the Koala video-LLM demo page! If you have any questions or feedback, please feel free to contact us. 
 Current online demo uses the 7B version of Llama-2 due to resource limitations.
 
 
 """
 
 Note_markdown = ("""
-### We note that our Global-Local QFormer model may be limited at understanding videos from rare domains. Due to the pretraining data, the \
+### We note that our Koala video-LLM model may be limited at understanding videos from rare domains. Due to the pretraining data, the \
     model may be susceptible to hallucinations
 We would like to acknowledge the Video-LLama repository which we copied the demo layout from.