@inproceedings{zhu-etal-2026-doclens,
    title = "{D}oc{L}ens: A Tool-Augmented Multi-Agent Framework for Long Visual Document Understanding",
    author = "Zhu, Dawei  and
      Meng, Rui  and
      Chen, Jiefeng  and
      Li, Sujian  and
      Pfister, Tomas  and
      Yoon, Jinsung",
    editor = "Liakata, Maria  and
      Moreira, Viviane P.  and
      Zhang, Jiajun  and
      Jurgens, David",
    booktitle = "Proceedings of the 64th Annual Meeting of the {A}ssociation for {C}omputational {L}inguistics (Volume 1: Long Papers)",
    month = jul,
    year = "2026",
    address = "San Diego, California, United States",
    publisher = "Association for Computational Linguistics",
    url = "https://preview.aclanthology.org/ingest-acl/2026.acl-long.1234/",
    pages = "26804--26829",
    ISBN = "979-8-89176-390-6",
    abstract = "Comprehending long visual documents, where information is distributed across extensive pages of text and visual elements, is a critical but challenging task for modern Vision-Language Models (VLMs). Existing approaches falter on a fundamental challenge: evidence localization. They struggle to retrieve relevant pages and overlook fine-grained details within visual elements, leading to limited performance and model hallucination. To address this, we propose DocLens, a tool-augmented multi-agent framework that effectively ``zooms in'' on evidence like a lens. It first navigates from the full document to specific visual elements on relevant pages, then employs a sampling-adjudication mechanism to generate a single, reliable answer. Paired with Gemini-2.5-Pro, DocLens achieves state-of-the-art performance on MMLongBench-Doc and FinRAGBench-V, surpassing even human experts. The framework{'}s superiority is particularly evident on vision-centric and unanswerable queries, demonstrating the power of its enhanced localization capabilities."
}