Metraipolis › archive › dc_cb62990183d4

BindCLIP: One Balanced Coupling For Compositional Vision Language Scoring — arXiv:2609.23717

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.23717
License (as archived)
CC0-1.0
Archived
2026-10-09T22:03:46.958Z
SHA-256 of the archived text
0a8f407ab8b6a78c832e42b6849fdedec3ee31572eaf2d30cf23e0dee877db89
Stable URL of this page
https://metraipolis.com/documento/dc_cb62990183d4

How to cite

BindCLIP: One Balanced Coupling For Compositional Vision Language Scoring — arXiv:2609.23717 — arXiv (en), https://arxiv.org/abs/2609.23717 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_cb62990183d4 (Metraipolis archive dc_cb62990183d4, sha256 0a8f407ab8b6a78c832e42b6849fdedec3ee31572eaf2d30cf23e0dee877db89)
BibTeX
@misc{dccb62990183d4,
  title = {BindCLIP: One Balanced Coupling For Compositional Vision Language Scoring — arXiv:2609.23717},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.23717},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_cb62990183d4; sha256 0a8f407ab8b6a78c832e42b6849fdedec3ee31572eaf2d30cf23e0dee877db89},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.23717. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Global vision--language similarities compress an image and a caption into one vector, preserving semantics but not which word corresponds to which region or how those regions are arranged; a model can recognize every word and object yet prefer a compositionally incorrect caption. We argue that a frozen encoder retains this association structure, so the problem is to read it rather than to rebuild it beside the pretrained similarity. We introduce BindCLIP, a pairwise scorer built on one latent object: a balanced token--patch--depth optimal-transport coupling that places both candidate captions and several visual depths in a single plan. Semantic, entity, order, and spatial evidence are read as energies of this state, and exchanging the candidates permutes the plan, making the score exactly antisymmetric. A geometric refinement inside the coupling contracts moves that the candidates and the visual depths do not support. No task label, parser, relation inventory, or detector is used. One checkpoint and one inference path improve the official What'sUp, ARO, and SugarCrepe benchmarks over frozen global CLIP, with the strongest transfer on the relation splits. Controls rule out patch access and caption-length shortcuts, and an inference-time lesion localizes spatial arrangement to the coupling.

Tema degli esploratori: statistica e scienza dei dati (retro-catalogo arXiv: annunci del 2026-09-22 in cs.CV).
Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV).
Pagina del lavoro: https://arxiv.org/abs/2609.23717 — 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 2026-10-09T21:45:20.945Z dalla pagina https://arxiv.org/catchup/cs.CV/2026-09-22?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).