Update README.md

README.md

@@ -20,15 +20,23 @@ dataset_info:
     dtype: string
   splits:
   - name: test
-    num_bytes: 479770102.0
+    num_bytes: 479770102
     num_examples: 217
   download_size: 473031413
-  dataset_size: 479770102.0
+  dataset_size: 479770102
 configs:
 - config_name: mantis_eval
   data_files:
   - split: test
     path: mantis_eval/test-*
+license: apache-2.0
+task_categories:
+- question-answering
+language:
+- en
+pretty_name: Mantis-Eval
+size_categories:
+- n<1K
 ---
 
-This is a newly curated dataset to evaluate multimodal language models' capability to reason over multiple images. More details are shown in https://tiger-ai-lab.github.io/Mantis/. 
+This is a newly curated dataset to evaluate multimodal language models' capability to reason over multiple images. More details are shown in https://tiger-ai-lab.github.io/Mantis/.