jdiazchao/seriality-gap

The Seriality Gap in Video Diffusion Models

Jorge Diaz Chao^*     Konpat Preechakul^*     Yuxi Liu     Yutong Bai

UC Berkeley
{jdiazchao,konpat,yuxi_liu,yutongbai}@berkeley.edu

Paper | Project Site | Code

 

 

Figure 1. Dependent-event prediction exposes the seriality gap. (a) Hard-sphere dynamics separate non-serial from serial video prediction. (i) In the single-ball control, any future state can be computed directly from the initial state, without resolving intermediate states. (ii) With multiple balls, each ball-ball collision changes the state governing later collisions, creating dependent-event chains that must be resolved in temporal order. (b) Given initial frames, can a diffusion model remain accurate as longer prediction horizons demand more serial computation?

Download a Model

Clone the seriality-gap GitHub repository, then follow its setup instructions.

git clone https://github.com/jdiazchao/seriality-gap
cd seriality-gap

Once setup is complete, download a checkpoint by passing its training configuration to download.sh:

bash download.sh configs/49f-5n/Train-20k-49f-5n-0c-B-768d-30l-64b-2e4lr.yaml

You may equivalently pass the model name directly:

bash download.sh 20k-49f-5n-0c-B-768d-30l-64b-2e4lr

Downloaded weights are saved to:

ckpt/huggingface/<model-name>/model-avg.safetensors

BibTeX

@misc{chao2026serialitygapvideodiffusion,
      title={The Seriality Gap in Video Diffusion Models}, 
      author={Jorge Diaz Chao and Konpat Preechakul and Yuxi Liu and Yutong Bai},
      year={2026},
      eprint={2607.13031},
      archivePrefix={arXiv},
      primaryClass={cs.LG},
      url={https://arxiv.org/abs/2607.13031}, 
}