Update README.md

README.md

@@ -10,7 +10,7 @@ pretty_name: VEC-Corpus
 
 ### Data Summary
 
-The model is trained on a Vietnamese text error correction dataset constructed from real-world noisy inputs. The dataset contains approximately 50,000 sentence pairs and is split into training, validation, and test sets.
+The model is trained on a Vietnamese text error correction dataset constructed from real-world noisy inputs. The dataset contains approximately 70,000 sentence pairs and is split into training, validation, and test sets.
 
 • **Data Source:** Crawled Vietnamese social media comments, reflecting informal and user-generated text.