Metraipolis › archive › dc_e8d520d20f13

GeoShrink: Accelerating Diffusion Transformers with Two Lines of Code — arXiv:2609.33723

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.33723
License (as archived)
CC0-1.0
Archived
2026-10-07T23:51:44.364Z
SHA-256 of the archived text
cd611d80fde88c8cbb461ccb902176705181376b1cf3aad40554833fa4e3ab1b
Stable URL of this page
https://metraipolis.com/documento/dc_e8d520d20f13

How to cite

GeoShrink: Accelerating Diffusion Transformers with Two Lines of Code — arXiv:2609.33723 — arXiv (en), https://arxiv.org/abs/2609.33723 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_e8d520d20f13 (Metraipolis archive dc_e8d520d20f13, sha256 cd611d80fde88c8cbb461ccb902176705181376b1cf3aad40554833fa4e3ab1b)
BibTeX
@misc{dce8d520d20f13,
  title = {GeoShrink: Accelerating Diffusion Transformers with Two Lines of Code — arXiv:2609.33723},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.33723},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_e8d520d20f13; sha256 cd611d80fde88c8cbb461ccb902176705181376b1cf3aad40554833fa4e3ab1b},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.33723. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Diffusion transformers incur substantial inference cost through repeated model evaluations along a sampling trajectory. We introduce GeoShrink, a training-free acceleration method that retains the original solver grid while evaluating the model only at a prescribed set of anchors. At skipped stages, GeoShrink predicts the solver-facing output by adding a geometrically retained fraction of the latest observed innovation to the most recent exact output. We derive this rule from chordal tangent transport and round-trip line projection, and establish a geometric anchor-spacing principle that minimizes the largest adjacent gap expansion under fixed coverage and first span. The analysis characterizes the geometric closure and propagation of prediction errors without assuming access to future model outputs. Experiments cover image, video, motion, and audio generation, together with adapted 3D backends. At approximately $5\times$ acceleration, GeoShrink improves FLUX PSNR by 3.10 dB over the strongest listed baseline. On HunyuanVideo, it achieves a reported $4.99\times$ speedup and improves ChronoMagic-Bench-150 PSNR by 5.44 dB over the strongest listed fidelity baseline. Comparisons at fixed evaluation budgets further show substantial gains on motion, audio, music, and 3D generation.

Tema degli esploratori: statistica e scienza dei dati (retro-catalogo arXiv: annunci del 2026-09-29 in cs.CV).
Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV).
Pagina del lavoro: https://arxiv.org/abs/2609.33723 — 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 2026-10-07T23:29:27.246Z dalla pagina https://arxiv.org/catchup/cs.CV/2026-09-29?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).