Add dataset card

README.md

@@ -22,6 +22,8 @@ dataset_info:
     dtype: string
   - name: patent_id
     dtype: string
+  - name: image
+    dtype: image
   - name: patent_title
     dtype: string
   - name: caption
@@ -40,29 +42,24 @@ dataset_info:
     dtype: int32
   - name: n_figures_in_patent
     dtype: int32
-  - name: year
-    dtype: int32
-  - name: svg
-    dtype: string
   - name: sibling_figure_ids
-    list: string
+    sequence: string
+  - name: reference_numerals
+    sequence:
+      struct:
+      - name: numeral
+        dtype: string
+      - name: label
+        dtype: string
   - name: locarno_class
     dtype: string
   - name: patent_date
     dtype: string
-  - name: image
-    dtype: image
+  - name: year
+    dtype: int32
   splits:
   - name: train
-    num_bytes: 621790
     num_examples: 500
-  download_size: 95097
-  dataset_size: 621790
-configs:
-- config_name: default
-  data_files:
-  - split: train
-    path: data/train-*
 ---
 
 # Patent Wireframes
@@ -114,11 +111,6 @@ Each record is a single patent figure paired with:
 - **Multi-view reasoning**: predict unseen views given other views
 - **Component understanding**: reference numeral grounding
 
-## Splits
-
-Train/val/test split at the **patent level** (not figure level) to prevent
-multi-view data leakage. 80/10/10 ratio, random with seed 42.
-
 ## Citation
 
 If you use this dataset, please cite: