SlideVQA: A Dataset for Document Visual Question Answering on Multiple Images.
Ryota TanakaKyosuke NishidaKosuke NishidaTaku HasegawaItsumi SaitoKuniko SaitoPublished in: CoRR (2023)
Keyphrases
- question answering
- multiple images
- passage retrieval
- information retrieval
- single image
- text summarization
- natural language
- information extraction
- multiple views
- candidate answers
- light source
- natural language processing
- named entities
- document classification
- document retrieval
- question classification
- qa clef
- visual features
- visual information
- syntactic information
- keywords
- cross language
- text documents
- question answering systems
- natural language questions
- answer validation
- document collections
- answer extraction
- relevance model
- relevant documents
- user queries
- qa systems
- query expansion
- co occurrence