DIS-CO/MovieTection
收藏Hugging Face2025-03-31 更新2025-02-15 收录
下载链接：
https://hf-mirror.com/datasets/DIS-CO/MovieTection
下载链接
链接失效反馈官方服务：
资源简介：
MovieTection数据集是一个为检测大型视觉语言模型（VLMs）训练数据中预训练数据的基准。它用于分析模型对版权视觉内容的接触情况。数据集包括14000个从100部电影中提取的帧及其对应的文本描述，用于图像/文本描述基础上的问题回答任务，模型需预测给定帧或其对应文本描述的电影名称。
The MovieTection dataset is a benchmark designed for detecting pretraining data in Large Vision-Language Models (VLMs). It serves as a resource for analyzing model exposure to Copyrighted Visual Content. The dataset includes 14,000 frames extracted from 100 movies along with their corresponding textual descriptions, used for image/text-based question-answering tasks where models are required to predict the movie title given a frame or its textual description.
提供机构：
DIS-CO
