microsoft/World-R1

World-R1是一个用于文本到视频世界模拟的提示数据集。它伴随论文《World-R1: Reinforcing 3D Constraints for Text-to-Video Generation》发布，其中使用强化学习来提高生成视频的3D一致性，同时保持视觉质量和运动多样性。数据集包含描述静态环境、动态场景和相机感知视频生成场景的英文提示。它设计用于在3D和运动约束下对文本到视频模型进行后训练、评估和分析。数据集提供两种配置：final和enhanced，每种配置包含不同的分割（train, test, dynamic）。数据集仅包含文本提示，不包含生成的视频、奖励注释、人类偏好标签或模型检查点。

World-R1 is a prompt-only dataset for text-to-video world simulation. It accompanies the paper World-R1: Reinforcing 3D Constraints for Text-to-Video Generation, where reinforcement learning is used to improve 3D consistency while preserving visual quality and motion diversity in generated videos. The dataset contains English prompts that describe static environments, dynamic scenes, and camera-aware video generation scenarios. It is designed for post-training, evaluation, and analysis of text-to-video models under 3D and motion constraints. The dataset provides two configurations: final and enhanced, each containing different splits (train, test, dynamic). The dataset includes text prompts only and does not include generated videos, reward annotations, human preference labels, or model checkpoints.