New DOSA framework enhances long document structure analysis

Researchers have developed a new framework called DOSA (DOcument Structure Analyzer) to improve the analysis of long and visually rich documents. This framework uses a tree-guided, self-regressive approach to infer hierarchical and ordering relationships between page objects like tables and headers. By fusing visual, textual, and layout features, DOSA incrementally builds a semantic tree that provides structural context for subsequent analysis, leading to significant improvements on benchmarks like DocHieNet. AI

IMPACT This framework could improve information retrieval and understanding in complex, multi-page documents.

RANK_REASON The cluster contains a research paper detailing a new framework for document structure analysis. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.AI →

• alphaXiv
• arXiv
• CatalyzeX Code Finder for Papers
• Connected Papers
• CORE Recommender
• DagsHub
• DocHieNet
• DOSA
• Gotit.pub
• Hugging Face
• Influence Flower
• Litmaps
• ScienceCast
• scite Smart Citations

• paper
• other

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →