English

EFlow: Fast Few-Step Video Generator Training from Scratch via Efficient Solution Flow

Computer Vision and Pattern Recognition 2026-03-31 v1

Abstract

Scaling video diffusion transformers is fundamentally bottlenecked by two compounding costs: the expensive quadratic complexity of attention per step, and the iterative sampling steps. In this work, we propose EFlow, an efficient few-step training framework, that tackles these bottlenecks simultaneously. To reduce sampling steps, we build on a solution-flow objective that learns a function mapping a noised state at time t to time s. Making this formulation computationally feasible and high-quality at video scale, however, demands two complementary innovations. First, we propose Gated Local-Global Attention, a token-droppable hybrid block which is efficient, expressive, and remains highly stable under aggressive random token-dropping, substantially reducing per-step compute. Second, we develop an efficient few-step training recipe. We propose Path-Drop Guided training to replace the expensive guidance target with a computationally cheap, weak path. Furthermore, we augment this with a Mean-Velocity Additivity regularizer to ensure high fidelity at extremely low step counts. Together, our EFlow enables a practical from-scratch training pipeline, achieving up to 2.5x higher training throughput over standard solution-flow, and 45.3x lower inference latency than standard iterative models with competitive performance on Kinetics and large-scale text-to-video datasets.

Keywords

Cite

@article{arxiv.2603.27086,
  title  = {EFlow: Fast Few-Step Video Generator Training from Scratch via Efficient Solution Flow},
  author = {Dogyun Park and Yanyu Li and Sergey Tulyakov and Anil Kag},
  journal= {arXiv preprint arXiv:2603.27086},
  year   = {2026}
}
R2 v1 2026-07-01T11:42:00.886Z