A-OKVQA: A Benchmark for Visual Question Answering using World Knowledge.
Dustin SchwenkApoorv KhandelwalChristopher ClarkKenneth MarinoRoozbeh MottaghiPublished in: CoRR (2022)
Keyphrases
- question answering
- world knowledge
- named entities
- natural language processing
- information extraction
- noun phrases
- natural language text
- background knowledge
- bag of words
- natural language
- relation extraction
- knowledge sources
- information retrieval
- visual features
- cross language
- visual information
- automatic extraction
- named entity recognition
- information access
- image retrieval
- low level