Update README.md

README.md

@@ -1,4 +1,4 @@
-Process reward models (mistral-7b) used in [Math-Shepherd](https://arxiv.org/pdf/2312.08935.pdf). 
+Process reward model (mistral-7b) used in [Math-Shepherd](https://arxiv.org/pdf/2312.08935.pdf). 
 
 `Input`: question + step-by-step solutions with a special step tag `ки`, e.g.,
 ```