Add image-text-to-text as task category

README.md

@@ -1,24 +1,24 @@
 ---
-license: apache-2.0
-task_categories:
-- visual-question-answering
-- question-answering
 language:
 - en
 - zh
+license: apache-2.0
 size_categories:
 - 1K<n<10K
-
+task_categories:
+- visual-question-answering
+- question-answering
+- image-text-to-text
 configs:
 - config_name: DynVQA_en
-  data_files: 
+  data_files:
   - split: test
-    path: "test/DynVQA_en/DynVQA_en.202412.jsonl"
+    path: test/DynVQA_en/DynVQA_en.202412.jsonl
   default: true
 - config_name: DynVQA_zh
-  data_files: 
+  data_files:
   - split: test
-    path: "test/DynVQA_zh/DynVQA_zh.202412.jsonl"
+    path: test/DynVQA_zh/DynVQA_zh.202412.jsonl
 ---
 
 # 📚 Dyn-VQA Dataset
@@ -53,4 +53,4 @@ The json item of Dyn-VQA dataset is organized in the following format:
       url={https://arxiv.org/abs/2411.02937}, 
 }
 ```
-When citing our work, please kindly consider citing the original papers. The relevant citation information is listed here.
+When citing our work, please kindly consider citing the original papers. The relevant citation information is listed here.