Metraipolis › archive › dc_0c431998fb8f

Spatially-Grounded Text-to-Video Generation via Inference-Time Gradient-Free Optimization — arXiv:2608.13037

Source
arXiv (en)
Original
https://arxiv.org/abs/2608.13037
License (as archived)
CC0-1.0
Archived
2026-10-19T04:55:43.054Z
SHA-256 of the archived text
62f0ad4be6274c8ef91e2512aa78d2c4e231e2ea66a5a1dde231880e58990306
Stable URL of this page
https://metraipolis.com/documento/dc_0c431998fb8f

How to cite

Spatially-Grounded Text-to-Video Generation via Inference-Time Gradient-Free Optimization — arXiv:2608.13037 — arXiv (en), https://arxiv.org/abs/2608.13037 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_0c431998fb8f (Metraipolis archive dc_0c431998fb8f, sha256 62f0ad4be6274c8ef91e2512aa78d2c4e231e2ea66a5a1dde231880e58990306)
BibTeX
@misc{dc0c431998fb8f,
  title = {Spatially-Grounded Text-to-Video Generation via Inference-Time Gradient-Free Optimization — arXiv:2608.13037},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2608.13037},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_0c431998fb8f; sha256 62f0ad4be6274c8ef91e2512aa78d2c4e231e2ea66a5a1dde231880e58990306},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.13037. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Diffusion Transformer Text-to-Video models have achieved remarkable synthesis quality, yet fine-grained spatial controllability remains a significant challenge. While existing training-free methods produce solid overall results in spatially grounded generation, \ie, placing a specific object in a designated location, they rely on gradient-based optimization techniques that incur prohibitive computational overhead, a bottleneck amplified in modern large-scale architectures. To address this limitation, we present Gradient-free Analytical Trajectory Optimization Video Generation (GATO-Vid), a novel training-free and gradient-free approach for precise spatial guidance. Rather than relying on costly backward passes, we introduce an alternative cross-attention score and solve it analytically to obtain an exact, closed-form solution. To use our analytical solution, we propose an on-the-fly injection mechanism tailored to the topological manifold of the transformer's latent space. Our experiments demonstrate that GATO-Vid significantly outperforms existing baselines in localization accuracy while introducing minimal computational overhead.

Tema degli esploratori: statistica e scienza dei dati (retro-catalogo arXiv: annunci del 2026-08-14 in cs.CV).
Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV).
Pagina del lavoro: https://arxiv.org/abs/2608.13037 — 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 2026-10-19T04:46:30.147Z dalla pagina https://arxiv.org/catchup/cs.CV/2026-08-14?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).