Datasets:

Modalities:
Text
Formats:
parquet
Languages:
Arabic
Size:
< 1K
License:
balhafni commited on
Commit
0ae1f3f
·
verified ·
1 Parent(s): a659f45

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +2 -0
README.md CHANGED
@@ -49,6 +49,8 @@ The corpus is designed to support research on sentence segmentation in Modern St
49
  AraSeg contains manually annotated documents collected from diverse sources and genres, enabling robust evaluation across different writing styles and domains.
50
  The benchmark contains manually annotated documents collected from diverse sources and genres, enabling robust evaluation across different writing styles and domains.
51
 
 
 
52
  ---
53
  ## Dataset Structure
54
 
 
49
  AraSeg contains manually annotated documents collected from diverse sources and genres, enabling robust evaluation across different writing styles and domains.
50
  The benchmark contains manually annotated documents collected from diverse sources and genres, enabling robust evaluation across different writing styles and domains.
51
 
52
+ AraSeg-PA is the Paragraph-Aware (PA) variant of the corpus where paragraph boundaries are provided.
53
+
54
  ---
55
  ## Dataset Structure
56