MichiganNLP/MAiDE-up

---
license: mit
task_categories:
- text-classification
- zero-shot-classification
language:
- en
- zh
- fr
- de
- it
- ko
- ru
- es
- tr
- ro
tags:
- misinformation
- LLM
- multilingual
- reviews
pretty_name: MAiDE-up
size_categories:
- 10K<n<100K
---

# Dataset Card for MAiDE-up: Multilingual Deception Detection of GPT-generated Hotel Reviews

## Table of Contents
- [Dataset Description](#dataset-description)
- [Dataset Summary](#dataset-summary)
- [Languages](#languages)
- [Supported Tasks and Leaderboards](#supported-tasks-and-leaderboards)
- [Additional Information](#additional-information)
- [Citation Information](#citation-information)
- [Contributions](#contributions)

## Dataset Description

- **Homepage:** [MAiDE-up homepage](https://github.com/MichiganNLP/multilingual_reviews_deception)
- **Repository:** [MAiDE-up repository](https://github.com/MichiganNLP/multilingual_reviews_deception)
- **Paper:** [Multilingual Deception Detection of GPT-generated Hotel Reviews](https://arxiv.org/abs/2404.12938)
- **Point of Contact:** [Oana Ignat](mailto:[e-mail])

### Dataset Summary

Multilingual Deception Detection of GPT-generated Hotel Reviews. We compare real hotel reviews from [Booking](https://Booking.com) with LLM-generated hotel reviews in 10 languages.

### Languages

The text in the dataset is in 10 languages: Chinese, English, French, German, Italian, Romanian, Korean, Russian, Spanish, Turkish

### Supported Tasks and Leaderboards
TODO

## Additional Information

### Citation Information

TODO

### Contributions

Thanks to [@oignat](https://github.com/OanaIgnat), [@snowphoebe](https://github.com/SnowPhoebe)

原始信息汇总

数据集卡片 MAiDE-up：多语言 GPT 生成酒店评论的欺骗检测

数据集描述

数据集概述

多语言 GPT 生成酒店评论的欺骗检测。我们比较了来自 Booking 的真实酒店评论与 10 种语言的 LLM 生成酒店评论。

语言

数据集中的文本包含 10 种语言：中文、英语、法语、德语、意大利语、罗马尼亚语、韩语、俄语、西班牙语、土耳其语。

支持的任务和排行榜

待定

附加信息

引用信息

待定

贡献

感谢 @oignat 和 @snowphoebe