Not All Tokens Need 40 Steps: Heterogeneous Step Allocation in Diffusion Transformers for Efficient Video Generation

arXiv 2026 20.8 method

TLDR

Proposes Heterogeneous Step Allocation (HSA), a training-free inference algorithm that assigns variable denoising steps to video DiT tokens for efficient video generation.

Reasoning

Strengths include a practical training-free approach with clear mechanisms (KV-cache synchronization and cached Euler updates) and strong empirical gains on multiple video generation models. Weaknesses visible from the abstract include limited detail on limitations, evaluation metrics, and comparison to broader world-model methods.

Read-first score

Read-first score 20.8, weighted from topical fit, citation, graph, method, reproducibility, and recency signals. Original total remains 5.

Recency 5%

100

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

Methodology quality 15%

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

Reproducibility 15%

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

Topical relevance 25%

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

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: 125.

Keyword Scores

video world model

world model

generative world model

world dynamics prediction

world simulator

interactive world model

model-based reinforcement learning world model

Resources

Paper