EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

SSM Meets Video Diffusion Models: Efficient Long-Term Video Generation with Structured State Spaces

Yuta Oshima, Shohei Taniguchi, Masahiro Suzuki, Yutaka Matsuo

New Generation Computing · 2026-08 · 44 (3)

DOI: 10.1007/s00354-026-00326-8 ↗

Kurz gefasst

Fachartikel in der Zeitschrift New Generation Computing, online erschienen am 27.06.2026. Verfasst von Yuta Oshima und 3 weiteren Autor:innen. Die Arbeit nennt 71 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Japan Society for the Promotion of Science London.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Given the remarkable achievements in image generation using diffusion models, the research community has shown increasing interest in extending these models to video generation. Recent diffusion models for video generation have predominantly utilized attention layers to extract temporal features. However, attention layers are limited by their computational cost, which increases quadratically with sequence length. This limitation poses significant challenges when generating longer video sequences using diffusion models. To overcome these challenges, we propose to leverage state-space models (SSMs) as temporal feature extractors. SSMs (e.g., Mamba) have recently garnered attention as promising alternatives owing to their linear-time memory and time consumption relative to the sequence length. Employing SSMs to capture temporal dependencies in video generation enables significantly higher generative performance at the same computational cost (e.g., memory usage, inference time) compared to attention-based methods, particularly for long-term sequences. For various model sizes, we comprehensively evaluated multiple long-term video datasets: MineRL Navigate, GQN-Mazes, and CARLA-Town01. For 256-frame video sequences, SSM-based models incur lower computational cost to achieve the same Fréchet Video Distance as attention-based models. Furthermore, the ablation study shows that when using SSMs for temporal modeling, incorporating bidirectionality and selective scans enhances video generation performance. Our code is available at https://anonymous.4open.science/r/SSM-Meets-Video-Diffusion-Models-067D/README.md .

Förderung

  • Japan Society for the Promotion of Science London

Zitieren

APA

Oshima, Y., Taniguchi, S., Suzuki, M., & Matsuo, Y. (2026). SSM Meets Video Diffusion Models: Efficient Long-Term Video Generation with Structured State Spaces. New Generation Computing, 44(3). https://doi.org/10.1007/s00354-026-00326-8

Vancouver

Oshima Y, Taniguchi S, Suzuki M, Matsuo Y. SSM Meets Video Diffusion Models: Efficient Long-Term Video Generation with Structured State Spaces. New Generation Computing. 2026;44(3). doi:10.1007/s00354-026-00326-8

BibTeX
@article{Oshima2026,
  author = {Oshima, Yuta and Taniguchi, Shohei and Suzuki, Masahiro and Matsuo, Yutaka},
  title = {SSM Meets Video Diffusion Models: Efficient Long-Term Video Generation with Structured State Spaces},
  journal = {New Generation Computing},
  year = {2026},
  volume = {44},
  number = {3},
  doi = {10.1007/s00354-026-00326-8}
}
Kostenlos starten →