Upload README.md with huggingface_hub

README.md

@@ -1,43 +1,4 @@
----
-dataset_info:
-  features:
-  - name: uid
-    dtype: string
-  - name: question
-    dtype: string
-  - name: answer
-    dtype: string
-  - name: question_type
-    sequence: string
-  - name: time_reference
-    dtype: string
-  - name: video_path
-    dtype: string
-  - name: video_info
-    struct:
-    - name: duration_minutes
-      dtype: float64
-    - name: fps
-      dtype: float64
-    - name: resolution
-      struct:
-      - name: height
-        dtype: int64
-      - name: width
-        dtype: int64
-  - name: key
-    dtype: string
-  - name: type
-    dtype: string
-  splits:
-  - name: train
-    num_bytes: 477238
-    num_examples: 1549
-  download_size: 177205
-  dataset_size: 477238
-configs:
-- config_name: default
-  data_files:
-  - split: train
-    path: data/train-*
----
+# LVBench Dataset
+
+This is a video question-answering dataset with subsets: cartoon, documentary, live, selfmedia, sport, and tv.
+