WeaQA: Weak Supervision via Captions for Visual Question Answering.
Pratyay BanerjeeTejas GokhaleYezhou YangChitta BaralPublished in: ACL/IJCNLP (Findings) (2021)
Keyphrases
- question answering
- visual features
- information retrieval
- information extraction
- question classification
- question answering systems
- cross language
- natural language processing
- passage retrieval
- natural language
- natural language questions
- open domain question answering
- sentence retrieval
- qa clef
- named entities
- low level
- visual information
- probabilistic model
- answer validation
- visual content
- image annotation
- video content
- syntactic information
- candidate answers
- data mining