README.md

---
license: cc-by-nc-sa-4.0
language:
- 'no'
---

# Dataset Card

## Dataset Summary

NO-CNN-DailyMail is a Norwegian news summarization dataset partially machine translated from English version of [CNN Dailymail Dataset](https://huggingface.co/datasets/cnn_dailymail?row=34). The summaries were written by journalists at CNN and the DailyMail. The dataset can be used for Machine reading comprehension and abstractive summarization tasks.

## Data Instances

For each instance, there is an *article* string and a *positive_sample* string representing news article and abstractive summary to this article.

## Data Split

The dataset is split into train and test sets.

|       | #samples |
|-------|---------------------|
| train | 61181              |
| test  | 15287               | 

For English version of this dataset, please refer to [link](https://paperswithcode.com/dataset/cnn-daily-mail-1)

## Citation Information
If you feel our work is helpful, please cite our paper:

```
@inproceedings{liu2024nlebench+,
  title={NLEBench+NorGLM: A Comprehensive Empirical Analysis and Benchmark Dataset for Generative Language Models in Norwegian},
  author={Liu, Peng and Zhang, Lemei and Farup, Terje and Lauvrak, Even and Ingvaldsen, Jon and Eide, Simen and Gulla, Jon Atle and Yang, Zhirong},
  booktitle={Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing},
  pages={5543--5560},
  year={2024}
}
```