NasrinImp/Twitter_AI

视觉对抗图灵测试（VCT²）数据集是一组用于评估和分析了人工智能生成图像检测（AGID）挑战的数据集，包含来自《纽约时报》Twitter账户的推文文本以及相应的真实图像和由多种最先进文本到图像模型生成的图像。每个样本包括一个推文文本、一个真实的Twitter图像，以及来自不同模型的生成图像，如Stable Diffusion 3.5、Stable Diffusion 3、Stable Diffusion 2.1、Stable Diffusion XL和DALL·E 3。该数据集共有约15909个样本，总大小约为10GB，支持在社交媒体环境中对真实性检测、语义一致性和人工智能与人类视觉相似度的研究。

The Visual Counter Turing Test (VCT²) dataset is a collection designed to benchmark and analyze the challenges of AI-generated image detection (AGID) using real-world, social media-driven captions and imagery. It includes tweet texts from The New York Times Twitter account along with corresponding real images and images generated by various state-of-the-art text-to-image models such as Stable Diffusion 3.5, Stable Diffusion 3, Stable Diffusion 2.1, Stable Diffusion XL, and DALL·E 3. Each sample comprises a tweet text, a real Twitter image, and generated images from different models. The dataset consists of approximately 15,909 examples with a total size of about 10GB, supporting research on authenticity detection, semantic consistency, and AI-human visual similarity in social media contexts.