BIFE: Better Interaction, Fewer Errors for Minute-Long Video Generation

arXiv 2025 25.9 method

TLDR

BIFE improves minute-long video generation using semantic sparse KV cache and Block Forcing training, introducing InterVBench and achieving state-of-the-art results.

Reasoning

The paper addresses important challenges in long video generation with novel mechanisms and a new benchmark, showing clear quantitative gains. However, the abstract only mentions world models as motivation, not as a core contribution, and lacks details on limitations or broader real-world deployment.

Read-first score

Read-first score 25.9, weighted from topical fit, citation, graph, method, reproducibility, and recency signals. Original total remains 5.

Recency 6%

86.7

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

Methodology quality 18%

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

Reproducibility 18%

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

Topical relevance 29%

7.1

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: 100.

Keyword Scores

world model

video world model

generative world model

world simulator

interactive world model

world dynamics prediction

model-based reinforcement learning world model

Resources

Paper