LongTake: Learning to Sustain Dynamics in Long-Horizon Video Generation

Key points

• Autoregressive (AR) video diffusion provides a natural framework for long-horizon generation, yet extended rollouts often become near-static or lose visual quality.1
• We hypothesize that these failures reflect the limited guidance provided by short-video supervision on how ongoing scene dynamics develops over longer durations.1
• This motivates us to introduce LongTake, a two-stage training pipeline built around Long-Horizon Teacher Forcing (TF) on curated real long videos.1
• Long-Horizon TF trains the AR model to predict later frames conditioned on long ground-truth video prefixes, extending direct supervision beyond the short training horizon.1

Sources (1)

• [1]LongTake: Learning to Sustain Dynamics in Long-Horizon Video Generation
  Hugging Face Daily Papers · Sep 29, 12:00 AM

  World models, game simulators, and long-take video creation require coherent scene evolution and sustained dynamics over extended durations.

  Autoregressive (AR) video diffusion provides a natural framework for long-horizon generation, yet extended rollouts often become near-static or lose visual quality.

Extractive summary: sentences quoted from the sources.