Update README.md

README.md

@@ -91,7 +91,7 @@ For a fair comparison, the questions for train and test are different.
 The results are shown below:
 
 | Source                          | Success of Parsing | Consistency |
-| ------------------------------- | ------------------ | ----------- |
+| :-----------------------------: | :----------------: | :---------: |
 | deepseek-coder-v2-lite-instruct | 100                | 95.08       |
 | deepseek-math-7b-RL             | 99.55              | 94.20       |
 | mathqwen-7b-Instruct            | 100                | 95.32       |