Stable Video Infinity: Infinite-Length Video Generation with Error Recycling

arXiv 2025 22.8 method

TLDR

Stable Video Infinity generates infinite-length videos by recycling errors via fine-tuning Diffusion Transformers to correct its own errors.

Reasoning

The paper introduces a novel error-recycling fine-tuning method to bridge the training-test discrepancy in autoregressive video generation, which is a strong conceptual contribution. However, the abstract lacks explicit mention of real-world experiments or benchmarks, and the core focus is on video generation, not world models, limiting relevance to the specified keywords.

Read-first score

Read-first score 22.8, weighted from topical fit, citation, graph, method, reproducibility, and recency signals. Original total remains 1.

Recency 8%

86.7

Uses a gentle age decay so recent papers surface without erasing older foundations. 2025

Methodology quality 25%

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

Reproducibility 25%

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

Topical relevance 42%

1.4

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

Field roles

Frontier

Rank sensitivity

Stability: volatile; rank range: 43.

Keyword Scores

video world model

world model

world simulator

generative world model

interactive world model

world dynamics prediction

model-based reinforcement learning world model

Resources

Paper