VideoScore2: Think before You Score in Generative Video Evaluation

📃Paper | 🌐Website | 💻GitHub | 🛢️Dataset | 🤗Model

VideoScore2 is a next-generation, interpretable and multi-dimensional video evaluation model designed to align with human judgment on text-to-video generation tasks.
It explicitly evaluates visual quality, text-to-video alignment, and physical/common-sense consistency, producing structured scores and reasoning.
Examples
Upload your video Text-to-Video Prompt

📚 Citation

@misc{he2025videoscore2thinkscoregenerative,
    title={VideoScore2: Think before You Score in Generative Video Evaluation}, 
    author={Xuan He and Dongfu Jiang and Ping Nie and Minghao Liu and Zhengxuan Jiang and Mingyi Su and Wentao Ma and Junru Lin and Chun Ye and Yi Lu and Keming Wu and Benjamin Schneider and Quy Duc Do and Zhuofeng Li and Yiming Jia and Yuxuan Zhang and Guo Cheng and Haozhe Wang and Wangchunshu Zhou and Qunshu Lin and Yuanxing Zhang and Ge Zhang and Wenhao Huang and Wenhu Chen},
    year={2025},
    eprint={2509.22799},
    archivePrefix={arXiv},
    primaryClass={cs.CV},
    url={https://arxiv.org/abs/2509.22799}, 
}