Generate then Select: Open-ended Visual Question Answering Guided by World Knowledge.
Xingyu FuSheng ZhangGukyeong KwonPramuditha PereraHenghui ZhuYuhao ZhangAlexander Hanbo LiWilliam Yang WangZhiguo WangVittorio CastelliPatrick NgDan RothBing XiangPublished in: CoRR (2023)
Keyphrases
- question answering
- open ended
- world knowledge
- knowledge sources
- noun phrases
- information extraction
- named entities
- natural language text
- natural language processing
- information retrieval
- natural language
- learning outcomes
- relation extraction
- bag of words
- cross language
- visual information
- background knowledge
- visual features
- low level
- image representation
- domain specific
- test collection
- visual data
- image classification
- artificial intelligence