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

arXiv 2026 22.7 method

TLDR

Introduces LongTake, a two-stage training pipeline with Long-Horizon Teacher Forcing on real long videos to sustain dynamics and visual quality in long-horizon autoregressive video diffusion.

Reasoning

The paper directly addresses a key limitation of autoregressive video diffusion—static degradation in long rollouts—with a simple training strategy and reports improved dynamic degree in 30-second outputs. However, the abstract lacks details on dataset scale, baseline comparisons, and evaluation metrics, limiting a fuller assessment.

Read-first score

Read-first score 22.7, weighted from topical fit, citation, graph, method, reproducibility, and recency signals. Original total remains 22.

Recency 5%

100

Uses a gentle age decay so recent papers surface without erasing older foundations. 2026

Topical relevance 25%

31.4

Uses existing LLM keyword relevance scores normalized to 0-100. world model,world simulator,generative world model,interactive world model,video world model,world dynamics prediction,model-based reinforcement learning world model

Methodology quality 15%

Screens visible abstract and analysis fields for experiment, dataset, baseline, metric, and limitation evidence. markers=baseline

Reproducibility 15%

Screens links and visible text for paper, code, dataset, artifact, and repository signals. pdf=True; code=False; dataset=False; markers=none

Graph prestige 15%

5.7

PageRank-style prestige over the local citation graph built from available OpenAlex references. local_pagerank=0.0016

Citation impact 15%

Uses OpenAlex-shaped citation metadata as a bibliometric attention signal, separate from paper quality. cited_by_count=0

Citation velocity 10%

Citation velocity estimates citations per publication-year to reduce old-paper bias. velocity=0.00

Field roles

Frontier

Rank sensitivity

Stability: volatile; rank range: 163.

Keyword Scores

video world model

world dynamics prediction

world model

generative world model

world simulator

interactive world model

model-based reinforcement learning world model

Resources

Paper