PropFly is a novel training pipeline for propagation-based video editing that eliminates the need for large-scale, paired (source and edited) video datasets. Instead, it leverages on-the-fly supervision from pre-trained Video Diffusion Models (VDMs).
Propagation-based video editing enables precise user control by propagating a single edited frame into subsequent frames while maintaining the original context. Our proposed method, PropFly, achieves this by:
This approach ensures temporally consistent and dynamic transformations, significantly outperforming state-of-the-art methods on various video editing tasks (evaluated on EditVerseBench and TGVE benchmarks).
1├── PropFly-1.3B/
2│ ├── diffusion_pytorch_model.bin # Model weights
3├── .gitattributes
4└── README.md
1@article{seo2026propfly,
2 title={PropFly: Learning to Propagate via On-the-Fly Supervision from Pre-trained Video Diffusion Models},
3 author={Seo, Wonyong and Moon, Jaeho and Lee, Jaehyup and Kim, Soo Ye and Kim, Munchurl},
4 journal={arXiv preprint arXiv:2602.20583},
5 year={2026}
6}
