Update README.md

README.md

@@ -51,7 +51,6 @@ This dataset was created by modifying and adapting the [allenai/QASPER: a datase
 and **aims to generate Question-Answer Pairs from the Abstract, Introduction of an NLP Paper**.
 
 
-## Dataset Details
 
 ### Dataset Description
 
@@ -62,15 +61,17 @@ and **aims to generate Question-Answer Pairs from the Abstract, Introduction of
 - validation : 211 rows
 - test : 320 rows
 
-
 - **Curated by:** [@UNIST-Eunchan](https://huggingface.co/UNIST-Eunchan)
-### Dataset Sources [optional]
+
+- 
+### Dataset Sources
+
+This data is made by applying and processing [allenai/qasper](https://huggingface.co/datasets/allenai/qasper)
 
 <!-- Provide the basic links for the dataset. -->
 
 - **Repository:** [allenai/qasper](https://huggingface.co/datasets/allenai/qasper)
-- **Paper [optional]:** [More Information Needed]
-- **Demo [optional]:** [More Information Needed]
+
 
 ## Uses
 
@@ -79,18 +80,16 @@ and **aims to generate Question-Answer Pairs from the Abstract, Introduction of
 - **Question-based Summarization**
 <!-- Address questions around how the dataset is intended to be used. -->
 
-[More Information Needed]
 
 ## Dataset Creation
 
 ### Curation Rationale
 
-<!-- Motivation for the creation of this dataset. -->
+Long Document Summarization datasets, especially those for Research Paper Summarization, are very limited and scarce. 
 
-[More Information Needed]
+We tweak the existing data to provide domains and QA pairs specific to NLP among Research Papers. 
 
-### Source Data
+We expect to be able to generate multiple QA pairs if we let the model sample through training. 
 
-This data is made by applying and processing [allenai/qasper](https://huggingface.co/datasets/allenai/qasper)
+We will release the fine-tuned model in the future.
 
-<!-- This section describes the source data (e.g. news text and headlines, social media posts, translated sentences, ...). -->