Update README.md

Author: DD-DuDa

README.md

@@ -4,7 +4,7 @@
 
 BitDecoding is a high-performance, GPU-optimized system
 designed to accelerate long-context LLMs decoding with a low-bit KV
-cache. Acheive more than **3x speedup** than FlashDecoding-v2.
+cache. Acheive more than **3x speedup** than Flash Attention v2.
 ![overview](imgs/overview.png)
 ![scheme](imgs/scheme.png)
 
@@ -59,4 +59,4 @@ If you find BitDecoding useful or want to use in your projects, please kindly ci
 ```
 
 ## Acknowledgement
-BitDecoding is inspired by many open-source libraries, including (but not limited to) [flash-attention](https://github.com/Dao-AILab/flash-attention/tree/main), [flute](https://github.com/HanGuo97/flute), [Atom](https://github.com/efeslab/Atom), [omniserve](https://github.com/mit-han-lab/omniserve).
+BitDecoding is inspired by many open-source libraries, including (but not limited to) [flash-attention](https://github.com/Dao-AILab/flash-attention/tree/main), [flute](https://github.com/HanGuo97/flute), [Atom](https://github.com/efeslab/Atom), [omniserve](https://github.com/mit-han-lab/omniserve).