Add LIBERO dataset for VLM reward modeling

README.md

@@ -19,12 +19,26 @@ dataset_info:
     sequence: float32
   - name: is_robot
     dtype: bool
+  - name: metadata
+    struct:
+    - name: demo_id
+      dtype: string
+    - name: file_path
+      dtype: string
+    - name: original_file
+      dtype: string
+    - name: scene
+      dtype: string
+    - name: trajectory_info
+      dtype: string
+    - name: trajectory_length
+      dtype: int64
   splits:
   - name: train
-    num_bytes: 2591761
+    num_bytes: 2658696
     num_examples: 500
-  download_size: 894248
-  dataset_size: 2591761
+  download_size: 857144
+  dataset_size: 2658696
 configs:
 - config_name: default
   data_files:

data/train-00000-of-00001.parquet

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d19ebb4b94c0a35197eb28cedd61c1bf603499f4bac0ac47b7c999eae4107912
-size 894248
+oid sha256:7a5926ea0e5de3e2b6273ea073a99d4982aab30dddd0ec025fe58b92b3383343
+size 857144