Metraipolis › archive › dc_31bbb682f2ef

Beyond Single Object: Learning 3D Relations with Large Language Models — arXiv:2608.15710

Source: arXiv (en)
Original: https://arxiv.org/abs/2608.15710
License (as archived): CC0-1.0
Archived: 2026-10-18T18:04:18.948Z
SHA-256 of the archived text: c7992b9af6383251f3024aa219219ef62307f9733061dec1c4cdb96e33c3d5ff
Stable URL of this page: https://metraipolis.com/documento/dc_31bbb682f2ef

How to cite

Beyond Single Object: Learning 3D Relations with Large Language Models — arXiv:2608.15710 — arXiv (en), https://arxiv.org/abs/2608.15710 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_31bbb682f2ef (Metraipolis archive dc_31bbb682f2ef, sha256 c7992b9af6383251f3024aa219219ef62307f9733061dec1c4cdb96e33c3d5ff)

BibTeX

@misc{dc31bbb682f2ef,
  title = {Beyond Single Object: Learning 3D Relations with Large Language Models — arXiv:2608.15710},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2608.15710},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_31bbb682f2ef; sha256 c7992b9af6383251f3024aa219219ef62307f9733061dec1c4cdb96e33c3d5ff},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.15710. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

We address a fundamental gap in 3D-LLMs: existing models focus on single-object/scene description, struggling with detailed, inter-object comparison. We propose a framework for detailed object-level reasoning across multiple objects with three components: (1) MO3D (Multi-Object in 3D), an instruction dataset requiring fine-grained multi-object comparison; (2) Multi-3DLLM, using a minimal Patch-Interaction Transformer (PIT) that models inter-/intra-object relationships while preserving local geometry; (3) Mini-apps, two application-driven benchmarks (Shape Mating, Change Captioning) that probe geometric understanding for practical use. Recent 3D-LLMs and 2D-VLMs perform poorly on these tasks, lacking both comparison-centric design and geometric awareness. In contrast, Multi-3DLLM trained on our mixture data learns geometric reasoning, surpasses all baselines on MO3D, and provides positive transfer to single-object classification.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-08-18 in cs.AI).
Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV) ; Artificial Intelligence (cs.AI); Computation and Language (cs.CL).
Pagina del lavoro: https://arxiv.org/abs/2608.15710 — 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 2026-10-18T16:29:14.045Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-08-18?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).