Skip to content

Dataset · saghi776

Labeled Text Corpus for AI Detection

saghi776/detectai-dataset

A large auto-labeled text dataset for training AI-generated content classifiers.

What's inside

Contains 516,672 text examples with labels for AI generation detection tasks. The data was automatically gathered from 60 different sources via a Cloudflare Workers pipeline. Language and file formats are not specified in the description.

What you can do with it

  • Train a binary classifier to distinguish machine-generated text from human-written text
  • Fine-tune encoders like RoBERTa or DeBERTa for anti-plagiarism and moderation tasks
Tasks
Text classification
Downloads this month
20.3K
Likes
7

How to load it

from datasets import load_dataset

ds = load_dataset("saghi776/detectai-dataset")