Update README.md

README.md

@@ -38,7 +38,7 @@ The figure below shows that open VLMs perform poorly on NutritionQA, even after
 To compute the metrics, please use the following script:
 
 ```python
-def scifi_score(question: str, target: str, prediction: str) -> bool:
+def score(question: str, target: str, prediction: str) -> bool:
     prompt = f"""You are tasked with evaluating a model's prediction for a question-answering task. Your goal is to determine if the model's prediction is correct based on its meaning, not necessarily exact wording.
 
 Here is the question that was asked: