datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
cs_squad-3.0
Dataset Card for Czech Simple Question Answering Dataset 3.0
This a processed and filtered adaptation of an existing dataset. For raw and larger dataset, see Dataset Source section.
Dataset Description
The data contains questions and answers based on Czech wikipeadia articles.
Each question has an answer (or more) and a selected part of the context as the evidence.
A majority of the answers are extractive - i.e. they are present in the context in the exact form. The… See the full description on the dataset page: https://huggingface.co/datasets/fewshot-goes-multilingual/cs_squad-3.0.EgyLaw-Squad
Egyption Law Squad
this dataset was made for Question Answering task about Egyption Law specially Personal Status Law (قانون الاحوال الشخصية )
About the Dataset
dataset was created for a Graduation Project in Computers and Artificial intellgence at Helwan University under supervisation from Dr.Ensaf Hossen
About Team
Abdelrahman Ahmed Hamdy
Shehab Gamal-elden
Mohsen Hisham Mohamed
Maya Ahmed Abdelsatar
Nancy Ahmed Mostafa
Nour Khaled Ali
structurized_squad
Structurized SQuAD Dataset
This dataset is derived from Rajpurkar/SQuAD dataset by applying the structurization process described in the "Enhancing LLM's Cognition via Structurization" (Liu et al., 2024) article.
To check the structurization process details: https://arxiv.org/abs/2407.16434
Construction process
For the structurization process, Llama-3.1-8B-Instruct and Llama-3.3-70B-Instruct models were used.
Some structurized context data may missing some answers or… See the full description on the dataset page: https://huggingface.co/datasets/martineden/structurized_squad.simple-squad
Dataset Card for Simple SQuAD
This is a simplified version of SQuAD v1 dataset.
Dataset Details
Dataset Description
The modifications made the original SQuAD v1 dataset:
Extract only the needed columns to train a seq2seq model (context, question and answer)
Preprocess the text to have only English letters and some basic punctuation marks.
Punctuation marks are separated from the adjacent words (i.e: hello! -> hello !)
Remove context length greater than 175… See the full description on the dataset page: https://huggingface.co/datasets/Monsef/simple-squad.