Metraipolis › archive › dc_3bf382b5352e

VisProve: Learning Typed Visual Provenance for Scientific Document Question Answering — arXiv:2611.10642

Source
arXiv (en)
Original
https://arxiv.org/abs/2611.10642
License (as archived)
CC0-1.0
Archived
2026-11-10T18:42:41.197Z
SHA-256 of the archived text
7e5287a7a36f8026d7f1ee4f1b9d5e0f0b4ccf75422eee01fc2a2507a93ca15b
Stable URL of this page
https://metraipolis.com/documento/dc_3bf382b5352e

How to cite

VisProve: Learning Typed Visual Provenance for Scientific Document Question Answering — arXiv:2611.10642 — arXiv (en), https://arxiv.org/abs/2611.10642 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_3bf382b5352e (Metraipolis archive dc_3bf382b5352e, sha256 7e5287a7a36f8026d7f1ee4f1b9d5e0f0b4ccf75422eee01fc2a2507a93ca15b)
BibTeX
@misc{dc3bf382b5352e,
  title = {VisProve: Learning Typed Visual Provenance for Scientific Document Question Answering — arXiv:2611.10642},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2611.10642},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_3bf382b5352e; sha256 7e5287a7a36f8026d7f1ee4f1b9d5e0f0b4ccf75422eee01fc2a2507a93ca15b},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2611.10642. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Visual citations show where an answer is supported, but not how the cited content contributes to it. In scientific document question answering, a model may read a value directly, compress several observations, or combine evidence to infer a conclusion. We study typed visual provenance, where a system generates an answer together with source regions and explicit support relations. We introduce ReViE, a dataset of 3,941 questions over complete scientific documents, with page, section, and region annotations labeled as Quotation, Compression, or Inference. We further propose VisProve, which augments joint answer and provenance generation with differentiable Spatial Evidence Reading and Evidence-Answer Relation Feedback. VisProve conditions next-token prediction on localized visual evidence and its estimated support relation through a gated norm-bounded residual. On the document-disjoint ReViE test set, VisProve improves Typed-F1 from 3.52 to 5.48 and Section-F1 from 45.99 to 53.24 over joint supervised fine-tuning with the same backbone, while Page-F1 changes only marginally. These results suggest that coupling visual grounding with support relation modeling mainly improves fine-grained provenance, while the low absolute Typed-F1 reveals a substantial gap between answer generation and precise visual evidence attribution.

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.AI, nuovi lavori del giorno).
Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV) ; Artificial Intelligence (cs.AI).
Pagina del lavoro: https://arxiv.org/abs/2611.10642 — 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 2026-11-10T18:37:37.598Z dalla pagina https://arxiv.org/list/cs.AI/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).