RIVA: A Pre-trained Tweet Multimodal Model Based on Text-image Relation for Multimodal NER.
Lin SunJiquan WangYindu SuFangsheng WengYuxuan SunZengwei ZhengYuanyi ChenPublished in: COLING (2020)
Keyphrases
- pre trained
- named entity recognition
- image features
- input image
- multi modal
- single image
- training data
- image classification
- text summarization
- high resolution
- information extraction
- image representation
- training examples
- named entities
- image matching
- lighting conditions
- machine learning
- audio visual
- graph cuts
- computer vision