GenKIE: Robust Generative Multimodal Document Key Information Extraction.
Panfeng CaoYe WangQiang ZhangZaiqiao MengPublished in: EMNLP (Findings) (2023)
Keyphrases
- information extraction
- web documents
- information retrieval
- text documents
- multi modal
- information retrieval systems
- text mining
- document collections
- natural language processing
- precision and recall
- text summarization
- generative model
- extracting meaningful
- document classification
- document images
- multimodal interaction
- unstructured documents
- topic models
- question answering
- unsupervised learning
- search engine
- structured data
- retrieval systems
- named entities
- data driven
- semi supervised
- partial occlusion
- free text
- named entity recognition
- knowledge discovery
- bayesian networks
- document analysis
- multimedia
- database