SceneGATE: Scene-Graph based co-Attention networks for TExt visual question answering.
Siwen LuoFeiqi CaoFelipe NunezZean WenJosiah PoonSoyeon Caren HanPublished in: CoRR (2022)
Keyphrases
- question answering
- information retrieval
- syntactic information
- textual entailment recognition
- text summarization
- visual data
- question classification
- information extraction
- natural language processing
- passage retrieval
- semantic parsing
- visual information
- text retrieval
- text mining
- semantic roles
- free text
- cross language
- natural language
- answer validation
- natural language questions
- answer extraction
- visual features
- named entities
- news video
- question answering systems
- qa clef
- low level
- video sequences
- keywords
- question answer pairs
- text documents
- qa systems
- textual entailment
- relational databases
- relation extraction
- semi supervised
- sentence retrieval
- answering questions
- information retrieval systems