Transformer Module Networks for Systematic Generalization in Visual Question Answering.
Moyuru YamadaVanessa D'AmarioKentaro TakemotoXavier BoixTomotake SasakiPublished in: CoRR (2022)
Keyphrases
- question answering
- answer extraction
- natural language processing
- question classification
- information retrieval
- syntactic information
- passage retrieval
- information extraction
- named entities
- visual information
- natural language
- cross language
- natural language questions
- qa clef
- question answering systems
- relation extraction
- visual features
- textual entailment recognition
- sentence retrieval
- open domain question answering
- qa systems
- candidate answers
- semantic roles
- answer validation
- co occurrence
- low level