Metraipolis › archive › dc_6cd8b40ebb64

Position Forcing: Self-Conditioning 3D Generation — arXiv:2610.10342

Source
arXiv (en)
Original
https://arxiv.org/abs/2610.10342
License (as archived)
CC0-1.0
Archived
2026-10-08T12:48:03.870Z
SHA-256 of the archived text
c695f83cdfb43b570bf3a8eb9ec15c39dcaf5f6611aea38dc3ed2c6b652d83cf
Stable URL of this page
https://metraipolis.com/documento/dc_6cd8b40ebb64

How to cite

Position Forcing: Self-Conditioning 3D Generation — arXiv:2610.10342 — arXiv (en), https://arxiv.org/abs/2610.10342 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_6cd8b40ebb64 (Metraipolis archive dc_6cd8b40ebb64, sha256 c695f83cdfb43b570bf3a8eb9ec15c39dcaf5f6611aea38dc3ed2c6b652d83cf)
BibTeX
@misc{dc6cd8b40ebb64,
  title = {Position Forcing: Self-Conditioning 3D Generation — arXiv:2610.10342},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2610.10342},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_6cd8b40ebb64; sha256 c695f83cdfb43b570bf3a8eb9ec15c39dcaf5f6611aea38dc3ed2c6b652d83cf},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.10342. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Recent single-stage 3D generative models commonly adopt VecSet representations, encoding 3D shapes as unordered sets of latent tokens. However, compared with two-stage methods that provide explicit positional guidance, these models must implicitly infer token positions throughout denoising, limiting their generation quality. We observe that, despite the absence of explicit positional conditioning, VecSet tokens retain recoverable spatial correspondences. Building on this observation, we propose Position Forcing, a position-based self-conditioning framework. During denoising, Position Forcing recovers token positions from the current clean latent estimate, quantizes them at progressively finer resolutions according to the denoising stage, and feeds the resulting positional encodings back into the diffusion Transformer. This progressively refined positional feedback provides spatial guidance at a granularity appropriate to each denoising stage, guiding shape generation along a coarse-to-fine trajectory and substantially improving generation quality without a separate position generation stage. Experiments demonstrate that Position Forcing achieves strong performance among single-stage 3D generative methods and outperforms several competitive multi-stage approaches.

Tema degli esploratori: statistica e scienza dei dati (elenco arXiv cs.CV, nuovi lavori del giorno).
Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV).
Pagina del lavoro: https://arxiv.org/abs/2610.10342 — 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 2026-10-08T12:18:15.484Z dalla pagina https://arxiv.org/list/cs.CV/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).