Mode Seeking meets Mean Seeking for Fast Long Video Generation

arXiv 2026 20.5 method

TLDR

Proposes a training paradigm combining mode-seeking and mean-seeking losses with a Decoupled Diffusion Transformer to generate minute-scale videos with local fidelity and long-range coherence.

Reasoning

Strengths: novel decoupling of local fidelity and long-term coherence, using limited long videos plus a frozen short-video teacher, and enabling few-step fast generation. Weaknesses: the abstract lacks quantitative details and explicit benchmark names, so empirical claims are not fully verifiable from the abstract alone.

Read-first score

Read-first score 20.5, weighted from topical fit, citation, graph, method, reproducibility, and recency signals. Original total remains 2.

Recency 6%

100

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

Methodology quality 18%

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

Reproducibility 18%

Screens links and visible text for paper, code, dataset, artifact, and repository signals. pdf=True; code=False; dataset=False; markers=github

Topical relevance 29%

2.9

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

Citation impact 18%

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

Citation velocity 12%

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

Field roles

Frontier

Rank sensitivity

Stability: volatile; rank range: 42.

Keyword Scores

video world model

world dynamics prediction

world model

world simulator

generative world model

interactive world model

model-based reinforcement learning world model

Resources

Paper