Upload folder using huggingface_hub

README.md

@@ -6,7 +6,7 @@ pipeline_tag: text-generation
 <div align="center">
 <h1>UltraGist for Llama-2-7b-chat</h1>
 
-[<a href="https://github.com/namespace-Pt/UltraGist">Github</a>]
+[<a href="https://arxiv.org/abs/2405.16635">Paper</a>] [<a href="https://github.com/namespace-Pt/UltraGist">Github</a>]
 </div>
 
 UltraGist is a context compression method can **flexibly**, **effectively**, and **efficiently** to handle various context lengths and compression ratios. We apply UltraGist on [Llama-2-7b-chat](https://huggingface.co/meta-llama/Llama-2-7b-chat-hf).
@@ -64,7 +64,3 @@ with torch.no_grad():
   print(f"Memory:           {compressed_memory[0][0].shape}")
   print("*"*20)
 ```
-
-## Citation
-```
-```