hexuan21/VideoFeedback_r2k_bad5k

该数据集包含两个配置：annotated和real。每个配置都包含文本提示、图像序列、视频链接以及多个评估指标（如视觉质量、时间一致性、动态程度、文本到视频对齐度和事实一致性）。此外，还包含对话数据，每个对话有来源和内容两个字段。数据集主要用于训练，annotated配置包含37,401个示例，real配置包含2,500个示例。

This dataset includes two configurations: annotated and real. Each configuration contains text prompts, image sequences, video links, and several evaluation metrics (such as visual quality, temporal consistency, dynamic degree, text-to-video alignment, and factual consistency). Additionally, it includes conversation data, with each conversation having fields for source and content. The dataset is primarily used for training, with the annotated configuration containing 37,401 examples and the real configuration containing 2,500 examples.