ITA: Image-Text Alignments for Multi-Modal Named Entity Recognition.
Xinyu WangMin GuiYong JiangZixia JiaNguyen BachTao WangZhongqiang HuangKewei TuPublished in: NAACL-HLT (2022)
Keyphrases
- multi modal
- named entity recognition
- text summarization
- information extraction
- multiple modalities
- named entities
- natural language processing
- uni modal
- text mining
- multi modality
- video search
- maximum entropy
- image retrieval
- single modality
- image content
- image representation
- input image
- named entity disambiguation
- segmentation method
- image classification
- image annotation
- information retrieval
- segmentation algorithm
- image collections
- semantic concepts
- low level
- high dimensional
- text documents
- co occurrence
- semi supervised
- artificial intelligence