Metraipolis › archive › dc_e8533438bbfc

VL-Dedup: Graded Multimodal Entity Resolution for Image-Text Data Deduplication — arXiv:2611.03080

Source: arXiv (en)
Original: https://arxiv.org/abs/2611.03080
License (as archived): CC0-1.0
Archived: 2026-11-03T11:42:46.215Z
SHA-256 of the archived text: 2302e1802e78f0b5d215c8f0d235734789132196811d1cf020fd06fb5a2169ec
Stable URL of this page: https://metraipolis.com/documento/dc_e8533438bbfc

How to cite

VL-Dedup: Graded Multimodal Entity Resolution for Image-Text Data Deduplication — arXiv:2611.03080 — arXiv (en), https://arxiv.org/abs/2611.03080 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_e8533438bbfc (Metraipolis archive dc_e8533438bbfc, sha256 2302e1802e78f0b5d215c8f0d235734789132196811d1cf020fd06fb5a2169ec)

BibTeX

@misc{dce8533438bbfc,
  title = {VL-Dedup: Graded Multimodal Entity Resolution for Image-Text Data Deduplication — arXiv:2611.03080},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2611.03080},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_e8533438bbfc; sha256 2302e1802e78f0b5d215c8f0d235734789132196811d1cf020fd06fb5a2169ec},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2611.03080. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Image-text corpora contain copies of one image, captions that describe one image in different ways, captions reused for different images, and captions that do not match their images. Each case calls for a different curation action. Deduplication and multimodal entity resolution (ER) only decide whether records are the same, and they are evaluated on labeled pairs. Yet pair scores do not tell which records end up in the same cluster or which useful captions get deleted. We therefore formulate image-text deduplication as graded multimodal ER, which decides visual identity, caption handling, and mismatches for each pair, and we evaluate it at three levels: pairs, clusters, and the retained set. Our method, VL-Dedup, groups records of the same visual entity into clusters and caps the size of each cluster. Within each cluster, a vision-language model (VLM) and CLIPScore decide which captions to keep and which to flag. We also build VL-DedupBench, with 50,000 COCO and SOP records and 66,788 labeled pairs. On it, a classifier that selects no false edge among the labeled pairs still joins millions of record pairs from different entities. Spending VLM calls on the pairs inside clusters, rather than on pairs with uncertain scores, gives better caption decisions from the same model. Even when almost all clusters are correct, keeping one record per cluster loses facts in many deletions, while the deletions of VL-Dedup lose much less information.

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/2611.03080 — 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 2026-11-03T11:06:37.614Z dalla pagina https://arxiv.org/list/cs.CV/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).