datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
majid-multitask-dataset
Majid Multi-task Dataset | مجموعه داده چندوظیفهای مجید
English Description 🇬🇧
This dataset provides Persian–English text for multi-task NLP: text classification and question answering, including subtasks such as sentiment analysis and toxicity detection.
Dataset Details
Curated by: Maid121232 (Majid)
Languages: Persian (fa), English (en)
License: Apache-2.0
Size: Small (< 10k samples)
Tasks: Text Classification, Question Answering
Dataset Structure
Files: train.csv… See the full description on the dataset page: https://huggingface.co/datasets/Maid121232/majid-multitask-dataset.esg-news-sentiment-multitaskCOLING-2025-GENAI-MULTIMultiTaskTWONB1-ds
Dataset Card: Multi-Task German Text Classification Dataset
Dataset Description
This dataset contains German text samples labeled for three tasks:
Fake News Detection (is_fake)
Hate Speech Detection (is_hate_speech)
Toxicity Detection (is_toxic)
Each entry in the dataset has binary or missing labels for the respective tasks:
0: Negative
1: Positive
-1: Not labeled for the task
The dataset is useful for training and evaluating models on multi-task learning objectives… See the full description on the dataset page: https://huggingface.co/datasets/Shivangsinha/MultiTaskTWONB1-ds.Multi-task-Dataset-Sample