Work

ALDEN: Reinforcement Learning for Active Navigation and Evidence Gathering in Long Documents

Year: 2026

Type: conference-paper

Abstract: While Vision-language models (VLMs) interpret text-rich images effectively, they struggle with reasoning across long, multi-page documents.We present Active Long-DocumEnt Navigation (ALDEN), a multi-turn reinforcement learning framework that fine-tunes VLMs as interactive agents capable of actively navigating long, visually rich documents rather than passive readers.ALDEN features a novel fetch action that allows direct page indexing, complementing the classic search action and better exploiting document structure.To ensure training efficiency and stability, we introduce a rule-based cross-level reward for dense supervision and a visual-semantic anchoring mechanism using dual-path KL-divergence constraints.We train ALDEN on a curated corpus built from open-source datasets, filtering out trivial samples and rewriting queries to incentivize multi-turn navigation and fetch usage.Empirically, ALDEN achieves state-of-the-art results on five long-document benchmarks, offering a more accurate and efficient path for long-document understanding.All of our code and datasets will be made publicly available at Github 1 .

Source: Proceedings of the conference - Association for Computational Linguistics. Meeting

DOI: 10.18653/v1/2026.acl-long.611

Authors: Tianyu Yang, Terry Ruas, Yijun Tian, Jan Philip Wahle, Daniel Kurzawe, Bela Gipp

Institutions: Niedersächsische Staats- und Universitätsbibliothek Göttingen, University of Göttingen, University of Notre Dame

Language: English

Topic: Multimodal Machine Learning Applications

Subfield: Computer Vision and Pattern Recognition

Field: Computer Science

Domain: Physical Sciences

Open Access status: hybrid

Funders: Niedersächsische Ministerium für Wissenschaft und Kultur, Deutsche Forschungsgemeinschaft, Volkswagen Foundation

Award: Digitale Geochemische DatenInfrastruktur (DIGIS) für GEOROC 2.0