Open-Ended Visual Question Answering by Multi-Modal Domain Adaptation.
Yiming XuLin ChenZhongwei ChengLixin DuanJiebo LuoPublished in: EMNLP (Findings) (2020)
Keyphrases
- multi modal
- question answering
- open ended
- domain adaptation
- video search
- cross domain
- labeled data
- natural language processing
- information extraction
- natural language
- sentiment classification
- learning outcomes
- information retrieval
- visual features
- audio visual
- visual information
- semi supervised
- transfer learning
- named entities
- document classification
- semi supervised learning
- high dimensional
- image annotation
- pairwise
- learning environment