test update

app.py

@@ -55,7 +55,7 @@ with gr.Blocks() as block:
             
             df = get_df()
             df['Date'] = 'unknown'
-            df2 = v3.get_df('Overall-V2')
+            df2 = v3.get_df('Overall')
             min_size2, max_size2 = get_size_range(df2)
 
             with gr.Row():
@@ -229,4 +229,5 @@ with gr.Blocks() as block:
                 gr.Markdown(SUBMIT_INTRODUCTION, elem_classes="markdown-text")
 
 
-block.launch(share=True)
+if __name__ == "__main__":
+    block.launch(share=True, ssr_mode=False)

report_score_v3.py

@@ -74,7 +74,7 @@ EXPERIMENTS = [
 # Define the datasets grouped by modality
 modality2dataset = {
     "audio": [
-        'SpeechCommand', 'UrbanSound8k', 'ESC-50', 'NSynth', 'CREMA-D', 'Clotho', 'SoundDescs', 'TUTSound', 'TUTSound(hard)', 'AVE', 'SpeechCOCO'
+        'SpeechCommands', 'UrbanSound8K', 'ESC-50', 'NSynth', 'CREMA-D', 'Clotho', 'SoundDescs', 'TUTSound', 'TUTSound(hard)', 'AVE', 'SpeechCOCO'
         ], 
     "image": [
         "ImageNet-1K", "N24News", "HatefulMemes", "VOC2007", "SUN397", "Place365", "ImageNet-A", "ImageNet-R", "ObjectNet", "Country211",

utils_v3.py

@@ -50,7 +50,7 @@ DATA_TITLE_TYPE_AG = BASE_DATA_TITLE_TYPE + \
                     ['number'] * len(TASKS_AG)
 
 TABLE_INTRODUCTION = """**MMEB**: Massive MultiModal Embedding Benchmark \n
-                        Models are ranked based on **Overall-V2**."""
+                        Models are ranked based on **Overall**(V3)."""
 TABLE_INTRODUCTION_I = """**I-CLS**: Image Classification, **I-QA**: (Image) Visual Question Answering, **I-RET**: Image Retrieval, **I-VG**: (Image) Visual Grounding \n
                         Models are ranked based on **Image-Overall**\n
                         **Models from the old V1 leaderboard are missing detailed scores of each dataset.