Update README.md

README.md

@@ -1,3 +1,39 @@
 ---
-license: cc-by-sa-4.0
+annotations_creators:
+- Jordan Painter, Diptesh Kanojia
+language:
+- en
+license:
+- cc-by-sa-4.0
+multilinguality:
+- monolingual
+pretty_name: 'Utilising Weak Supervision to create S3D: A Sarcasm Annotated Dataset'
+size_categories:
+- 100K<n<1M
+source_datasets:
+- original
+task_categories:
+- text-classification
 ---
+
+## Table of Contents
+- [Dataset Description](#dataset-description)
+
+- 
+# Utilising Weak Supervision to Create S3D: A Sarcasm Annotated Dataset
+This is the repository for the S3D dataset published at EMNLP 2022. The dataset can help build sarcasm detection models.
+
+# S3D-v2 Summary
+The S3D-v2 dataset is our silver standard dataset of 100,000 tweets labelled for sarcasm using weak supervision by a majority voting system of fine-tuned sarcasm detection models. The models used are 
+our [roberta-large-finetuned-SARC-combined-DS](https://huggingface.co/surrey-nlp/roberta-large-finetuned-SARC-combined-DS), [bertweet-base-finetuned-SARC-DS](https://huggingface.co/surrey-nlp/bertweet-base-finetuned-SARC-DS)
+and [bertweet-base-finetuned-SARC-combined-DS](https://huggingface.co/surrey-nlp/bertweet-base-finetuned-SARC-combined-DS).
+
+S3D contains 13016 tweets labelled as sarcastic, and 86904 tweets labelled as not being sarcastic.
+# Data Fields
+
+- Label: A label to denote if a given tweet is sarcastic
+
+# Data Splits
+- Train: 70,000
+- Valid: 15,000
+- Test: 15,000