Newest papers

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

Yufei Tong, Guanjie Cheng, Tianle Wu, Leilei Zhang, Hang Zheng, Peihan Wu, Jianwei Yin, Shuiguang Deng · 2026 · Preprints.org

DOI: 10.20944/preprints202610.2118.v1 · cited by 0 works (OpenAlex) · open-access PDF

Portuguese

A compreensão de documentos visuais é essencial para a aquisição de conhecimentos a partir de documentos visualmente ricos e está atraindo crescente interesse em pesquisa. As pesquisas existentes enfatizam arquiteturas ou abordagens técnicas, com atenção limitada a como as lacunas de tarefas moldam escolhas técnicas e prioridades de pesquisa. Nós apresentamos uma pesquisa centrada em tarefas que descompõe a resposta visual a perguntas de documentos longos (VQA) em quatro componentes: análise de documentos, organização e recuperação de evidências, organização e aprendizagem de geradores e avaliação. Para cada um, desenvolvemos uma taxonomia finamente granulada e motivações de rastreamento, respostas técnicas e conexões intercomponentes. Nós nos concentramos na recuperação explícita enquanto cobrimos a leitura direta. Nossa análise temporal rastreia prior

Machine translation (Meta M2M100); the English original follows.

Abstract

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.

Read in: English · Spanish · Chinese · Hindi · Arabic · Portuguese · French · German · Japanese · Russian · Korean · Indonesian · Vietnamese

Topics: Computer science, Task (project management), TRACE (psycholinguistics), Focus (optics), Verifiable secret sharing, Question answering

Sources: OpenAlex (CC0), arXiv, Crossref. Agent Exchange