Metraipolis › archive › dc_b9fffd253fdb

Accelerating Video Diffusion via Training-Free Trajectory Routing — arXiv:2609.30096

Source: arXiv (en)
Original: https://arxiv.org/abs/2609.30096
License (as archived): CC0-1.0
Archived: 2026-10-08T04:28:01.750Z
SHA-256 of the archived text: b5aa4a597b66b53168619bcab95f2e235d123fb4a4a662d776d0fdc84ce783d5
Stable URL of this page: https://metraipolis.com/documento/dc_b9fffd253fdb

How to cite

Accelerating Video Diffusion via Training-Free Trajectory Routing — arXiv:2609.30096 — arXiv (en), https://arxiv.org/abs/2609.30096 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_b9fffd253fdb (Metraipolis archive dc_b9fffd253fdb, sha256 b5aa4a597b66b53168619bcab95f2e235d123fb4a4a662d776d0fdc84ce783d5)

BibTeX

@misc{dcb9fffd253fdb,
  title = {Accelerating Video Diffusion via Training-Free Trajectory Routing — arXiv:2609.30096},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.30096},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_b9fffd253fdb; sha256 b5aa4a597b66b53168619bcab95f2e235d123fb4a4a662d776d0fdc84ce783d5},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.30096. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Video diffusion is computationally expensive, as it requires executing a large model across many denoising steps. Even with step-distillation, inference remains expensive because every distilled step still requires a costly model evaluation. We present TRACK: TRajectory-Aware Capacity routing via top-K selection, a heterogeneous denoising strategy that switches between compatible large and small models at selected steps, reducing the average cost per denoising evaluation. The switching steps are determined using a calibration process. TRACK first rolls out a reference trajectory with the large model. Then at each step, the small model's prediction is also collected and compared against the large model's prediction to obtain a relative disagreement score. Both models receive the same latent, timestep, conditioning, and guidance inputs. Aggregating this signal over a calibration set produces a disagreement score map across diffusion steps, which determines a switching policy for an efficient inference process: quality-sensitive steps keep using the large model, while steps with low disagreement scores are routed to the small model. Inference executes only the selected model at each step, requiring no retraining, architecture or scheduler changes, or online dual-model evaluation. Across Wan 2.1, Cosmos 3, TurboDiffusion, and FastVideo, TRACK yields $1.95\times$, $2.04\times$-$2.73\times$, $2.69\times$, and $2.17\times$ speedups, respectively, with comparable aggregate quality and high diversity retention. TRACK thereby establishes automated, training-free model switching as a practical acceleration paradigm for video diffusion.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-25 in cs.AI).
Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV) ; Artificial Intelligence (cs.AI).
Pagina del lavoro: https://arxiv.org/abs/2609.30096 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito).
Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina.
Letto il 2026-10-08T03:50:58.237Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-09-25?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).