A Vietnamese Dataset for Text Segmentation and Multiple Choices Reading Comprehension