Update utils.py

utils.py

@@ -39,7 +39,9 @@ This comprehensive suite enables robust evaluation of multimodal embedding model
 | [**🤗Hugging Face**](https://huggingface.co/datasets/TIGER-Lab/MMEB-V2) |
 """
 
-TABLE_INTRODUCTION = """Models are ranked based on V1-Overall."""
+TABLE_INTRODUCTION = """I-CLS: Image Classification, I-QA: (Image) Visual Question Answering, I-RET: Image Retrieval, I-VG: (Image) Visual Grounding \n
+V1-Overall = (10 * I-CLS + 10 * I-QA + 12 * I-RET + 4 * I-VG) / 36 \n
+Models are ranked based on V1-Overall."""
 
 LEADERBOARD_INFO = """
 ## Dataset Summary