Visually Guided Generative Text-Layout Pre-training for Document Intelligence.
Zhiming MaoHaoli BaiLu HouLifeng ShangXin JiangQun LiuKam-Fai WongPublished in: NAACL-HLT (2024)
Keyphrases
- visually guided
- cf loadingtexthtml
- text documents
- information retrieval
- web documents
- document analysis
- keywords
- textual content
- page layout
- digital documents
- text content
- document content
- text classifiers
- document layout
- scientific papers
- multimedia documents
- document images
- scientific documents
- document processing
- text mining
- generative model
- training corpus
- structured documents
- training set
- document set
- automatic text summarization
- semantic information
- document structure
- latent semantic analysis
- document image retrieval
- text collections
- text categorization
- topic models
- printed documents
- text corpus
- retrieval systems
- text lines
- document retrieval
- document corpus
- document type
- obstacle avoidance
- text retrieval
- text summarization
- discriminative classifiers
- electronic documents
- document level
- vector space model
- knowledge base