Preprint
Jul 2026
HierDoc: Hierarchical Page-to-Region Evidence Routing for Long-Document Visual Question Answering
HierDoc, a hierarchical evidence-routing framework that formulates long-document evidence acquisition as two-stage set prediction from pages to regions, achieves state-of-the-art or competitive performance among open-weight systems, improving LongDocURL by 16.87% relative to the strongest reported open-weight baseline.
Rongjian Gu, Weng Zhou, Junyu Xiong et al.
· 0 citations