Update README.md

README.md

@@ -121,7 +121,7 @@ Since the compared PRMs have not been trained on code-related tasks, this sectio
 | Reward Model             | Method                  | MBPP  | MBPP+ | HumanEval | HumanEval+ | LiveCodeBench-2024.01-2024-11 |
 |--------------------------|-------------------------|-------|-------|-----------|------------|-------------------------------|
 | N/A                      | Greedy Sampling Pass@1 | 81.7  | 69.3  | **84.8**  | **78.0**   | 25.3                          |
-| Skywork-o1-Open-PRM-7B   | Best-of-N@64           | **84.9** | **72.5** | 83.5      | **78.0**       |                              |
+| Skywork-o1-Open-PRM-7B   | Best-of-N@64           | **84.9** | **72.5** | 83.5      | **78.0**       | **30.7**                             |
 
 #### Llama3.1-8B-Instruct