Is GPT-3 all you need for Visual Question Answering in Cultural Heritage?
Pietro BonginiFederico BecattiniAlberto Del BimboPublished in: CoRR (2022)
Keyphrases
- question answering
- cultural heritage
- digital libraries
- multimedia
- information extraction
- question classification
- natural language processing
- natural language questions
- visual information
- question answering systems
- information retrieval
- passage retrieval
- cross language
- named entities
- low level
- visual features
- semantic roles
- digital archives
- syntactic information
- sentence retrieval
- natural language
- multi modal
- answer extraction
- open domain question answering
- answer validation
- qa clef
- relation extraction
- candidate answers
- machine learning
- relational databases
- textual entailment recognition