Metraipolis › archive › dc_e094e1cbd41d

A Task-Centric Survey of Long-Document Visual Question Answering: Taxonomy, Evolution, and Future Directions — arXiv:2610.31385

Source: arXiv (en)
Original: https://arxiv.org/abs/2610.31385
License (as archived): CC0-1.0
Archived: 2026-10-25T08:36:45.956Z
SHA-256 of the archived text: 38134b516952e352db3244b47c5eac967220dd83001c5ee06c9bcdb613d8a52f
Stable URL of this page: https://metraipolis.com/documento/dc_e094e1cbd41d

How to cite

A Task-Centric Survey of Long-Document Visual Question Answering: Taxonomy, Evolution, and Future Directions — arXiv:2610.31385 — arXiv (en), https://arxiv.org/abs/2610.31385 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_e094e1cbd41d (Metraipolis archive dc_e094e1cbd41d, sha256 38134b516952e352db3244b47c5eac967220dd83001c5ee06c9bcdb613d8a52f)

BibTeX

@misc{dce094e1cbd41d,
  title = {A Task-Centric Survey of Long-Document Visual Question Answering: Taxonomy, Evolution, and Future Directions — arXiv:2610.31385},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2610.31385},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_e094e1cbd41d; sha256 38134b516952e352db3244b47c5eac967220dd83001c5ee06c9bcdb613d8a52f},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.31385. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Visual document understanding is essential for acquiring knowledge from visually rich documents and is attracting growing research interest. Existing surveys emphasize architectures or technical approaches, with limited attention to how task bottlenecks shape technical choices and research priorities. We present a task-centric survey that decomposes long-document visual question answering (VQA) into four components: document parsing, evidence organization and retrieval, generator organization and learning, and evaluation. For each, we develop a fine-grained taxonomy and trace motivations, technical responses, and cross-component connections. We focus on explicit retrieval while covering direct reading. Our temporal analysis traces expanding priorities from multimodal content understanding to sufficient, high-fidelity evidence acquisition and sustained evidence management. We outline future directions in evidence sufficiency, verifiable and correctable understanding, and budget-aware coordination, with an outlook on continual document understanding.

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.CL, nuovi lavori del giorno).
Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV) ; Computation and Language (cs.CL); Information Retrieval (cs.IR).
Pagina del lavoro: https://arxiv.org/abs/2610.31385 — 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 2026-10-25T08:31:27.841Z dalla pagina https://arxiv.org/list/cs.CL/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).