CogniSAL/MDAIGT

MDAIGT（多领域人工智能生成文本检测）数据集是为了支持跨多个领域（包括新闻和学术写作）的人工智能生成英文文本检测研究而设计的。该数据集包含了人类编写的和机器生成的样本，旨在帮助评估AI文本检测模型在鲁棒性和领域泛化方面的表现。每个样本都被标记为“人类编写”或“机器生成”，为检测不同写作风格和上下文中的AI生成内容提供了一个宝贵的基准。

MDAIGT (Multi-Domain AI-Generated Text Detection) is a dataset designed to support research on detecting AI-generated English text across multiple domains, including news and academic writing. The dataset comprises both human-written and machine-generated samples, aiming to help evaluate the robustness and domain generalization of AI-text detection models. Each sample is labeled as human or machine, providing a valuable benchmark for detecting AI-generated content in different writing styles and contexts.