Merge pull request #195 from junming-yang/leaderboard

gen_table.py

@@ -63,7 +63,7 @@ def BUILD_L1_DF(results, fields):
     check_box = {}
     check_box['essential'] = ['Method', 'Parameters (B)', 'Language Model', 'Vision Model']
     # revise there to set defualt dataset
-    defualt_dataset = ['MMBench_TEST_EN', 'MMStar', 'MME', 'MMMU_VAL', 'MathVista', 'OCRBench', 'MMVet']
+    defualt_dataset = ['MMBench_TEST_EN', 'MMBench_TEST_CN', 'MMStar', 'MME', 'MMMU_VAL', 'MathVista', 'OCRBench', 'MMVet', 'AI2D', 'HallusionBench', 'LLaVABench', 'SEEDBench_IMG']
     check_box['required'] = ['Avg Score', 'Avg Rank'] + defualt_dataset
     check_box['avg'] = ['Avg Score', 'Avg Rank']
     check_box['all'] = check_box['avg'] + fields

meta_data.py

@@ -22,7 +22,7 @@ META_FIELDS = ['Method', 'Parameters (B)', 'Language Model', 'Vision Model', 'Op
 MAIN_FIELDS = [
     'MMBench_TEST_EN', 'MMBench_TEST_CN', 'MMStar', 'MME',
     'MMMU_VAL', 'MathVista', 'OCRBench', 'AI2D', 
-    'HallusionBench', 'SEEDBench_IMG', 'MMVet', 'LLaVABench'
+    'HallusionBench', 'SEEDBench_IMG', 'MMVet', 'LLaVABench', 'CCBench', 'RealWorldQA'
 ]
 MMBENCH_FIELDS = ['MMBench_TEST_EN', 'MMBench_DEV_EN', 'MMBench_TEST_CN', 'MMBench_DEV_CN', 'CCBench']
 MODEL_SIZE = ['<10B', '10B-20B', '20B-40B', '>40B', 'Unknown']